UNIVERSITY OF 
 ILLINOIS LIBRARY 
 URBANA-CHAMPAIGN 
 
 STACKS 
 
LIBRARY AUTOMATION 
 
 AS A SOURCE OF 
 MANAGEMENT INFORMATION 
 
 Papers presented at the 1982 Clinic on Library Applications 
 of Data Processing, April 25-28, 1982 
 
Clinic on Library Applications 
 of Data Processing: 1982 
 
j V 
 
 C, a. 
 
 Library Automation 
 
 as a Source of 
 Management Information 
 
 Edited by 
 F. WILFRID LANCASTER 
 
 Graduate School of Library and Information Science 
 University of Illinois at Urbana-Champaign 
 
Library of Congress Cataloging in Publication Data 
 
 Clinic on Library Applications of Data Processing (19th : 
 1982 : University of Illinois at Urbana-Champaign) 
 Library automation as a source of management infor- 
 mation. 
 
 "Papers presented at the 1982 Clinic on Library 
 Applications of Data Processing, April 25-28, 1982" Half 
 title p. 
 
 Includes index. 
 
 1. Library administration Data processing Con- 
 gresses. 2. Libraries Automation Congresses. 
 3. Library science Data processing Congresses. 
 I. Lancaster, F. Wilfrid. II. University of Illinois at 
 Urbana-Champaign. Graduate School of Library and 
 Information Science III. Title. 
 
 Z678.C63 1982 025.3'028'54 83-9110 
 
 ISBN: 0-87845-068-8 
 
 1983 The Board of Trustees of The University of Illinois 
 
CONTENTS 
 
 Introduction 1 
 
 F. WILFRID LANCASTER 
 
 Management, Information and the Organization: 
 
 Homily from the Experience of the Data Rich 
 
 but Information Poor 2 
 
 RUSSELL SHANK 
 
 Tutorial on Management Information Systems 10 
 
 RICHARD J. ROLAND, JR. 
 
 Beyond the Numbers A Decision Support System 27 
 
 KEN DOWLIN 
 LYNN MAGRATH 
 
 Organizational Considerations Relating 
 to the Implementation and Use of 
 
 Management Information Systems 59 
 
 KATHLEEN M. HEIM 
 
 Geac as a Source of Management Information 72 
 
 WAYNE MULLIN 
 
 Characteristics of Managerial Resistance 
 
 to Library Management Information Systems 92 
 
 JOHN N. OLSGAARD 
 
 Management Information Systems 
 
 in a Network Environment Ill 
 
 MARY ELLEN JACOB 
 
 NEAL K. KASKE 
 
 Computer-Derived Management Information 
 
 in a Special Library 128 
 
 ROBERTA. KENNEDY 
 
 Management Information from 
 
 Bibliographic Databases 148 
 
 JANE L. CALDWELL 
 
An Online General Ledger System 160 
 
 EDWIN B. BROWNRIGG 
 
 A Library Management Information System 
 
 in a Multi-Campus Environment 164 
 
 GLYN T. EVANS 
 
 ALBERT BEILBY 
 
 Index 197 
 
Introduction 
 
 The nineteenth annual Clinic on Library Applications of Data Processing 
 was held at the Illini Union, University of Illinois, April 25-28, 1982. 
 
 In the last twenty years, two of the most significant developments in 
 the library field have been the increase in the adoption of automated 
 procedures and growing interest in the measurement and evaluation of 
 library services. Yet, the marriage of these two trends that is, the use of 
 automated systems as sources of data to permit improved management and 
 decision-making has not been a major focus of professional interest. The 
 1982 clinic sought to correct this situation by focusing attention on the 
 management information component of automated systems in libraries. 
 As far as the organizers are aware, this was the first conference to deal in 
 toto with this particular topic. 
 
 The papers reproduced here represent great variety ranging from a 
 tutorial on management information and decision support systems 
 (Boland), through more philosophical discussions (Heim, Olsgaard and 
 Shank) on the value to the library manager of computer-derived informa- 
 tion, to studies of the use of automated systems as sources of management 
 data in libraries and information services of various types (Brownrigg, 
 Caldwell, Dowlin, Evans, Jacob/Kaske, Kennedy, Mullin). 
 
 This volume, we believe, contributes a significant body of literature 
 on an important topic that has received little previous attention in the 
 library profession. 
 
 F.W. LANCASTER 
 
 Editor 
 
RUSSELL SHANK 
 
 University Librarian and Professor 
 University of California at Los Angeles 
 
 Management, Information and the 
 
 Organization: Homily from the Experience 
 
 of the Data Rich but Information Poor 
 
 Calvin Mooers's Law has long been one of my favorites. I often use it as an 
 excuse for irrational behavior in library management. Over twenty years 
 ago, Calvin Mooers commented: "An information retrieval system will 
 tend not to be used whenever it is more painful and troublesome for a 
 customer to have information than for him not to have it." 1 
 
 Mooers was involved with information storage and retrieval systems 
 in science, but his words are just as appropriate for management informa- 
 tion systems (MIS). The problem with having information in Mooer's view 
 is that you can't just let it sit there if it indicates that something must be 
 done. If that something is hard to do or involves difficult social consequen- 
 ces (such as disrupting the faculty's habits of library use, or the student's 
 timing of meeting his date), it will only cause ulcers, sleepless nights or 
 unemployment if you don't make the adjustment. Of course there is one 
 way to avoid all of the trouble, and that is not to have the information in 
 the first place. 
 
 If Mooers's Law is right, then managers and administrators should 
 reject management information systems. Yet here one is, deliberating 
 positively on the future of such systems presumably enhanced to new levels 
 of capabilities by the computer. Well, I suppose that's all right, for in spite 
 of Mooers's Law or anything I might say, human nature makes one 
 inquisitive so that counting things and arraying the tally in various 
 displays comes with his genes. 
 
 But Mooers's Law alone is not sufficient. I have discovered Russell's 
 inference: "If there are two things to be counted, that which is the easiest 
 will cost the most to tally." Then there is Shank's syndrome: "If you count 
 one of two things the boss will want the other." This is followed closely by 
 
MANAGEMENT, INFORMATION to THE ORGANIZATION 3 
 
 the invariable derivation: "That which the boss wants is the most ambigu- 
 ous to define, the most expensive to count, but the most relevant to him." 
 And we have the petulant ukase: Never wait for the information do it 
 now and then gather data to prove you had the right idea but the "system" 
 subverted the purity of your solution and for irrelevant reasons. Or the 
 invariable proposition: The sum of the columns is never equal to the sum 
 of the footings. Or the perfect postulate: Don't just stand there count! 
 Why do we collect data which can be turned into management infor- 
 mation? Well, there can be many reasons: 
 
 1 . The data might be worth something. The choice of tasks among several 
 alternatives for the use of one's time, assuming that the utility of the 
 alternatives is the same, might be clearly indicated by the data which 
 shows how much resource would be required to perform each task. 
 
 2. We collect some data to sustain tradition. For fifty years research li- 
 braries have been reporting the size of their holdings and it has now 
 become a ritual to contribute to the ARL census. Some await the report 
 eagerly which year by year shows them bobbing up and down amongst 
 their peers in a variety of measurable elements. About the best I can say 
 of it is that it might sustain some ego gratification. Unless one is a 
 Harvard graduate there is always something to aim at. Or one can take 
 pleasure or not about the company he's in. 
 
 3. Some collecting is done just because the data is there. It is human nature 
 to sort things into piles, and to wander idly over the field counting 
 things as one goes. 
 
 4. Some data is collected so as to avoid embarrassment. Someone'mightask 
 questions about the operations which can best be answered with 
 numbers. What manager would be considered qualified for the job if he 
 or she couldn't tell how much of anything goes into or comes out of the 
 operation. There is, of course, a certain peace of mind which is derived 
 from collecting data, even if no one ever asks about it. At least one has 
 data which can be used in press releases. 
 
 5. There is, of course, the great public demand for accountability. Since 
 most of the public does not have the faith any more to believe the litanies 
 of those who spend public money, they need data to explain the use of 
 public money. We may snow them with numbers, but at least the 
 appearance of precision on our part will give the appearance that we 
 think we know what we are doing and we're doing it for them! 
 
 6. Sometimes data is collected in order to create tables and reports to over- 
 whelm the administration. Perhaps this is just another form of account- 
 ability. But busy administrators don't need more information (which by 
 the way might be accurate and painful to deal with). What one actually 
 finds is that less and less data is forwarded through the organization 
 
RUSSELL SHANK 
 
 since less and less is read or heeded. From one standpoint, therefore, the 
 function of an organization seems to be to suppress data. What is often 
 lacking is a distinction between data which contains information to 
 control processes at the local level, and data which explains the reason 
 for, or the need for, decisions on resource allocation to the next level. 
 The same data does not always provide the foundation for the ex traction 
 of information needed at each level. Hence the need to stratify the 
 management information system to match the management levels in an 
 organization. This would tend to make the notion of a standard set of 
 data to be gathered for the whole organization irrational. Furthermore, 
 the higher one goes in an organization, the more difficult it is to predict 
 what issues have to be handled, hence the more difficult it is to predict 
 management's need for information. 
 
 7. Data is collected so that those who give it those who are doing the 
 work think that managers know what's going on and care. There are 
 many signals which can be given after they get the data which will 
 sustain this impression. Part of the art of management is giving signals. 
 
 8. Data is collected because we cannot lose by doing it. Institutions assess 
 no penalties on departments including management information 
 departments for collecting too much data. Data gathering is seldom 
 priced, hence the evaluation of management information systems is 
 seldom based on the economics of running the systems. Management 
 requires data which it assumes can be gathered easily in the course of 
 doing the tasks being counted, and that takes us back to the human 
 nature theory of data gathering. 
 
 Just as there is no penalty for collecting too much data, there is no 
 premium on brevity of data collection. Information management 
 departments or systems are not judged on the basis of the value of the 
 decisions which could be based on the information in the data they 
 gather. Managers further up in the hierarchy are judged on the quality 
 of their decisions, which could possibly be based on information from a 
 management information system, but which does not necessarily need 
 to be. The whole system seems to be put into place as a perpetual motion 
 machine all too of ten installed without there being any analysis of what 
 to do with the data. 
 
 Now for any and all of these reasons, and probably more, we collect 
 and report data. But a typical administrative characteristic seems to be the 
 appearance of ignoring irrefutable information derived from good data. 
 This is particularly true in the political arena. There the black and white of 
 logic is overwhelmed by back scratching and other heuristic devices. Most 
 of our political and some of our administrative behavior seems to be based 
 on a sample of one preferably apocryphal incident. 
 
MANAGEMENT, INFORMATION b THE ORGANIZATION 5 
 
 If we are so good why is our data rejected up the line? My inventory of 
 reasons is long and sound, based on many years both as a producer of data 
 about operations, and as a manager whose primary functions seems to be 
 finding reasons why managers, administrators and politicians reject data. 
 This may occur because: 
 
 1. It is not clear what, if anything, can be done about whatever the data 
 purports to show. Here I have reworded Mooers's Law. Management 
 information will be rejected to the extent that it indicates that some- 
 thing has to be done. Fewer managers have been fired for doing nothing 
 than for trying something. 
 
 2. Rejection may occur because the data is not trusted. One way to reject a 
 whole report is to find one error in it. This error can be used to discredit 
 the whole report. Then the difficult decision can be avoided with 
 impugnity. 
 
 3. Rejection may occur if it is not clear if the data indicates something 
 good or something bad. Clear data, and good information, does little to 
 clarify ambiguous social conditions. 
 
 4. A corollary to the previous pronouncement is that data is rejected be- 
 cause those who present it do not have the theatrical skills required to 
 convince the hierarchy that it means anything. 
 
 5. Data is rejected because the recipient of the information is inclined to see 
 "the other side" of the issue. This is the side for which one did not gather 
 data. This is particularly true of social and political issues. The use of 
 this technique is one of the signs of the successful politician or 
 administrator. 
 
 6. Rejection may occur because the recipient has his own MIS which pro- 
 duces contrary data, or provides grounds for alternate inferences. This 
 produces enough ambiguity so that the manager or politican can dis- 
 play the wisdom of quick, seat-of-the-pants judgment. If the person has 
 been successful to date (that's why he or she is in the management 
 position), he or she will probably be right again. 
 
 7. Data is rejected because there is no way to match the data from different 
 but related departments or situations. This moves us again into the 
 arena of ambiguity and the need for a carefully designed, stratified 
 management information system. 
 
 8. Data rejection occurs because there is a lack of understanding along the 
 way as to what data and information will sustain decisions about the 
 value of services. This is either the result of a manager's inability to 
 direct the enterprise by thinking of scenarios for a possible future, or a 
 lack of empathy for the manager's problems by those down the line. 
 Lack of empathy may short-circuit as many upwardly mobile people as 
 it does the data they provide. 
 
RUSSELL SHANK 
 
 Perhaps I can be a bit more sanguine about management information 
 systems with the involvement of computers. Obviously if data storage and 
 retrieval is the name of the game, computers are ideally suited to this 
 function. They can tally and summarize and they can work fast. If the 
 summaries can be made hierarchical, just like the organization, then 
 computers can take over the screening (or data suppression) function but 
 with a difference. With a computer and the right software, the manager can 
 get not only his or her own compressed summary data, but at any time go 
 into the file and see anyone's data should the fancy strike him. 
 
 But I worry. Data system operators tell me that the cost of storage is 
 coming down rapidly has been for years. What this means is that it will 
 not cost so much on a unit basis to store the garbage; so for the money, we 
 can store more. But all that we will then have is cheaper garbage. 
 
 Computers have still another advantage. They can count the transac- 
 tions as they work on them. No longer does someone have to tally the 
 number of circulation cards at the end of the day, or the orders sent out, or 
 the number of items processed. Programmed correctly, the computer will 
 do the counting as it goes. Just look at the number on the bottom line. If 
 nothing else, this is a labor saver. 
 
 To me the chief advantage of the computer, though, is that it can 
 economically work on the "what if" questions. These are the ones that 
 managers like to ask, but usually cannot get answered because there is no 
 way to test all of the variables in order to substantiate answers without 
 operating on the patient (the library or its users) without an anesthetic. 
 
 The "what if" questions I am thinking of are operational. Adminis- 
 trators may ponder notions such as: What if we closed seven of the branches 
 two hours earlier each day, paid the staff to travel to a regional installation 
 for the remaining time, beefed up the reference staff for half the hours lost, 
 and added 25 percent to the budget of the book delivery system? What 
 would be the differential cost (excess or saved), and what would be the effect 
 on service (given some reasonable measures of service output of the 
 library)? This is not the kind of question that can be answered with 
 assurance. The hot, highly successful managers might be able to handle it. 
 After all, they are hot and highly successful because they have the intuition, 
 the experience, or the ability to make their analyses sound good, which 
 leads them to answer questions while dodging issues. 
 
 But in the main they tread gingerly take one step at a time try to 
 find those things which can be done relatively cheaply, and which do not 
 involve an intolerable sunk cost. If we have to retreat we at least want to do 
 it gracefully and cheaply. 
 
 The computer could handle all of the data for the mix of options we 
 would like to test if only. And here is the crux of the issue of making the 
 computer a real management ally. One could do it best if only he had good 
 
MANAGEMENT, INFORMATION b THE ORGANIZATION 7 
 
 models of the library as an operating organism, and lots of data about the 
 units of operation in the system. Then he could run libraries as models 
 with different variables, look at paper outputs and do it all without 
 disturbing current operations. Then one could make decisions if he had 
 confidence and a lot of courage. Intuition and experience might still be the 
 sine qua non of the administrator, but now the data output by the manage- 
 ment information system would be in the context of the management 
 questions. 
 
 Modeling and collecting a large volume of operational data would 
 then make some sense. The work of people such as Hamburg, Kantor and 
 King are steps in this direction. So is that of those who have done the many 
 user studies. But their work only provides proofs of the utility of the 
 process. They show the way to analyze problems and to count data. What 
 we need to do now is to build models of library operations. And the data to 
 be inserted in them must be specific to the library being analyzed. But 
 beware of the monster we might create. According to the Wall Street 
 Journal the major computer centers in the United States produced over 240 
 billion pages of printout in 1980 about 1000 pages forevery man, woman 
 and child. This number could grow to 10,000 pages per person by 1985 at 
 
 . o 
 
 the rate we are going. 
 
 Wildavsky said that: "Alas, access to data does not automatically 
 convert itself into information. Inference and interpretation are 
 required." 3 This is where administrative talent (whether based on intui- 
 tion or experience) must come to the fore. It cannot be replaced by the 
 computer. It could even be stymied by receiving thousands of pages, or 
 even thousands of lines, of data. Even computer graphics which coordinate 
 and display data in compact and different ways might not help. Some users 
 are visually illiterate. Here it may be well to note an even deeper issue. Not 
 only is there widespread visual illiteracy, but also there may be an even 
 greater antipathy if not illiteracy to mathematics. If computers lead to 
 an increase in quantitative reports we might expect an even greater rejec- 
 tion of the output of MIS by administrators and politicians. Unless we can 
 improve peoples' ability in general to handle numbers, the rejection syn- 
 drome is likely to be reinforced. 
 
 Areas for concern when the computer is brought into MIS abound. 
 While the data storage costs per bit are coming down rapidly, Parkinson's 
 Law prevails. Data will expand to fill storage space and the total cost of 
 the system will go up both because it costs a lot to collect all of the data that 
 can be stored, and because the actual cost of the storage mechanism goes up 
 even though it densely packs in a lot more data with each machine 
 generation. 
 
 Both the power of the computer and the cost of using it might force us 
 at last to pay the kind of attention to management information systems and 
 
RUSSELL SHANK 
 
 problems that has been notably lacking in our profession. We are 
 extremely cost conscious in every other aspect of library work. We establish 
 budgets for various phases of our operations. We seldom do it, however, for 
 management information systems. Budgets for management information 
 ought to be developed, and done so on the basis of the utility of the 
 information the systems can extract and give out. Or perhaps managers 
 should be given information budgets either in terms of cash they can 
 transfer to the management information system, or cash they can give for 
 information, or in terms of the time they are allowed to spend examining 
 data and information. (In a way, most administrators have a time budget. 
 Some management information systems output is rejected simply because 
 managers are enervated by merely seeing a pile of printout and wondering 
 where they will get the time to look at it.) 
 
 At last we may have the impetus to place heavier emphasis on the 
 education of managers to analyze and state issues in terms which will 
 suggest their information needs. And perhaps we can even learn to evaluate 
 administrators and managers for their ability to state issues and to make 
 inferences based on sound information derived from reliable and sensible 
 data. In this realm we are asking people to live symbolically. George Miller 
 of Princeton warns that: "More and more people will become useless if 
 they cannot live at the symbolic level." 4 The success of computer model- 
 ing, therefore, might well be proportional to the ability to live 
 symbolically. 
 
 We must, of course, recognize that the high technology of computers 
 does not mean that they are infallible. That is and this ought to be 
 obvious from what I have said so far people, not computers, solve prob- 
 lems. Neat columns, multi-inverted matrices, accurate tallies, quick eating 
 and consolidation of lots of data are not substitutes for intuition and 
 inference. 
 
 In case you missed it, here's where I have been. As a manager I am not 
 overwhelmingly enthusiastic about the utility of management informa- 
 tion systems up to now. I am skeptical about why we count, and I am not 
 certain we can see the way to use information derived from the data. We 
 have created reasons or perhaps just allowed human nature to take its 
 course to reject data. The computer, however, might finally get us nearer 
 Nirvana in the management information arena. A big task for the profes- 
 sion is to find ways to let the computer stimulate the organization under 
 different circumstances. Then maybe the manager's knowledge and intui- 
 tion will be supported with something more than faith. 
 
 But there is still work to do. The computer has to be fed and kept on 
 the right track. We could fail. I think of the story of the two men who were 
 cast adrift in a life boat in the cold North Atlantic Ocean. Just as they were 
 
MANAGEMENT, INFORMATION b THE ORGANIZATION 9 
 
 about to take their last breaths, one looked up and said: "Praise the Lord, 
 we're saved. Here comes the Titanic." 
 
 ACKNOWLEDGMENT 
 
 I am indebted to Aaron Wildavsky (UC Berkeley) whose views on management 
 information have been eloquently expressed in many places, but particularly in his 
 unpublished paper "Information as an Organizational Problem." 
 
 REFERENCES 
 
 1. Mooers, Calvin N. "Mooers 1 Law, or Why Some Retrieval Systems are Used and 
 Others arc Not." American Documentation ll(July 1960):ii. 
 
 2. Needle, David. "Managers Face a New Problem: An Information Glut." Infoworld 
 4(1 March 1982): 17. 
 
 3. Wildavsky, Aaron. "Information as an Organizational Problem." Paper presented at 
 a colloquium sponsored by Public Not-for-Profit Management and Computers and Infor- 
 mation Systems Curriculum Areas, Graduate School of Management, 9 Oct. 1981, UCLA. 
 
 4. Miller, George quoted in Fiske, Edward B. "Schools Enter the Computer Age: An 
 Analysis." New York Times, 25 April 1982, p. 38. 
 
RICHARD J. ROLAND, JR. 
 
 Associate Professor 
 
 Department of Accountancy 
 
 University of Illinois at Urbana-Champaign 
 
 Tutorial on Management 
 Information Systems 
 
 Introduction 
 
 Management information is communication that leads to managerial 
 action, and managerial action is a betterment achieved through a process 
 of planning and control. A critical distinction in discussing management 
 information is the difference between data and information. Data are any 
 coded messages, considered apart from their use by an individual. Informa- 
 tion, on the other hand, is the meaning of data to an individual. Informa- 
 tion, therefore, is derived from data through interpretation and is 
 ultimately a subjective phenomenon available only to the individual 
 interpreter. 
 
 The first problem in discussing management information is to resolve 
 the issue of subjectivity so that we might proceed with the question of 
 system design. The field of management information systems has done this 
 by treating as information, data that have been selectively assembled and 
 structured so that we believe they will be useful to their recipient because 
 we can adequately anticipate the meaning that will be gained. 
 
 Thus a discussion of management information always presupposes a 
 recipient, a context and a use. The recipient is an individual manager in an 
 organizational context who is engaged in decision-making activity. Man- 
 agement information must always be considered in light of all three of 
 these aspects: 
 
 1 . The decision situation includes an analysis of the type of problem being 
 addressed, the adequacy of evidence required, and the range of norma- 
 tive and descriptive models available for understanding how the deci- 
 sion is or should be made. 
 
 10 
 
MIS TUTORIAL 11 
 
 2. The organizational context includes an analysis of the structure, style, 
 climate, and power that characterizes the organization as well as aspects 
 of the larger culture that impact the norms of perception, cognition, 
 evaluation, and behavior used by its members. 
 
 3. The individual manager includes an analysis of the limits of cognitive 
 ability, the dynamics of group processes, and the cognitive style that 
 characterizes the way an individual collects and processes data. 
 
 Fundamentally, therefore, a discussion of management information 
 cannot be value free. If we are to move beyond a discussion of mere data, 
 then we must either affirm the status quo or propose a change in the 
 decision situation, the organizational context or the individual manager. 
 Either way, we take a normative position with respect to these three aspects 
 when we make a design statement about an information system. 
 
 An appreciation of the distinction between data and information leads 
 to a second problem in discussing management information. Data, as data, 
 have a cost; and data, as information, have a value. In general, the cost of 
 data increases as the amount collected increases, but the value of informa- 
 tion does not. Information has a marginal rate of return which diminishes 
 as its quantity increases. When we discuss the use of automation to replace 
 existing manual processes, we can identify reductions in the cost of labor, 
 space, time, etc. to produce a given output. Cost and value calculations, 
 though often imprecise, can be made. 
 
 However, when we move beyond using automation to process repeti- 
 tive transactions at the operational level, and explore the use of automa- 
 tion to enhance management decision-making, our ability to quantify the 
 value of an information system becomes very problematic. We shift away 
 from an assessment of efficiency and quantitative improvements toward an 
 assessment of effectiveness and qualitative improvements in the function- 
 ing of the organization. 
 
 The two problems of data versus information and cost versus value set 
 the stage for presenting a framework for analyzing management decision- 
 making and a process for developing information systems to support 
 decision-making. This paper is in four sections. Section one defines the 
 organization as a system and the manager as a decision-maker about that 
 system. Section two applies the framework to a library as a system. Section 
 three derives implications for information system design, and section four 
 explores the process of system development. 
 
 A Systems View of Organization 
 
 The systems approach is a broad label for the general attempt to 
 understand organizations by analyzing their relational and dynamic 
 
12 RICHARD J. BOLAND, JR. 
 
 aspects. The organization is viewed as a set of relationships between 
 component parts that stand apart from an environment, receiving inputs 
 from it and producing outputs that are received by it. Thus the organiza- 
 tion is an open system that is dependent on input and output relations with 
 its environment, and organizes its internal components to meet those input 
 and output demands. 
 
 Internally, each component of the organization is understood as the 
 relationship between its own subunits as it receives input from, and 
 produces output to other components of the organization or its environ- 
 ment. Thus, each level of analysis of the system (organization, component, 
 subunit, etc.) is both a whole with relations between subunits that must 
 be maintained and a part of a larger whole, with input and output 
 relations between other parts of that whole. 
 
 Systems are seen as the nested, hierarchical organization of relatively 
 self-contained sets of relationships between internal components, interact- 
 ing by input/output processes with a larger environment. This view is 
 important because it emphasizes that any "problem" with the organiza- 
 tion must be understood in terms of its internal and external relations. It 
 cannot be understood in isolation or out of context. Also, since organiza- 
 tions produce myriad outputs and are constituted by a very large number of 
 relationships, the role of managers as active determinants of the problems 
 they face (by defining inputs, outputs and relations of interest to them) is 
 made apparent. 
 
 The basic building block of the system approach is the notion of 
 input, process and output (see figure 1). 
 
 Environment 
 
 Input 
 
 Process 
 
 Output 
 
 Fig. 1. Basic Input-Process-Output Model 
 
 The process can be left as an unexplained "black box," or can be 
 expanded to include any level of detail of boxes within boxes. A very crude 
 application to a library would be as indicated in figure 2. 
 
 As a first elaboration of this crude image, we will add the concept of 
 levels of decision-making. The decision-maker can be viewed as making 
 strategic, managerial or operational level decisions. At a strategic level, the 
 
MIS TUTORIAL 
 
 13 
 
 Funds 
 Personnel 
 Clients 
 Publications 
 Other Resources 
 
 Access and 
 -^\ use of 
 
 holdings 
 
 Fig. 2. Application of the Input-Process-Output Model 
 
 decision-maker is concerned with defining and prioritizing goals and 
 objectives, and with securing the resources to achieve them. Here the 
 manager questions and refines the basic mission of the organization 
 what client's are emphasized, what type and scope of services are provided, 
 what will the character of the holdings be, what will the criteria for success 
 and performance evaluation be? 
 
 At the managerial level, the decision-maker takes as given the resour- 
 ces available, the statement of mission and priorities, and the standard of 
 performance evaluation. The problem is to arrange the operations, sched- 
 ule activities, and allocate resources for the purpose of effectively achieving 
 the strategic goals. The key idea here is effectiveness in the way the 
 organization is configured, and the way resources are allocated. 
 
 At the operational level, the decision-maker is concerned with the 
 details of procedures for carrying out organizational functions defined at 
 the managerial level. Here the emphasis is on efficiency in performance, on 
 reducing bottlenecks in flows through the system, and on removing unnec- 
 essary redundancy. 
 
 These three levels of decision-making are added to the basic input- 
 process-output model in figure 3. Each level of decision can be further 
 characterized by three stages of the decision-making process: intelligence, 
 design and choice. 
 
 Intelligence. This is the initial stageof a decision process in which the 
 manager is concerned with understanding the situation as a basis for 
 defining the need for action or identifying the need for decision-making. 
 The emphasis is on defining problems, threats, opportunities, and con- 
 straints that require action. 
 
 Design. At this stage, the decision-maker has identified a decision 
 problem, and is inventing alternative courses of action and developing 
 ideas for dealing with the problem. The recent emphasis on creativity in 
 
14 
 
 RICHARD J. BOLAND, JR. 
 
 Strategic Relations with Environment 
 
 Managerial 
 Effectiveness 
 
 Operational 
 Efficiency 
 
 Input 
 
 Library System: 
 
 Component 
 Procedures 
 Communications 
 Flows 
 
 Output 
 
 Fig. 3. Diagram of Decision-making Levels Applied to the 
 Basic Input-Process-Output Model 
 
 management training attests to the need for more attention to this stage of 
 the decision process. 
 
 Choice. This stage is frequently discussed as if it comprised the whole 
 of management decision-making. Here, a course of action is selected from 
 the set of alternatives that have been identified for meeting the needs of the 
 problem, as it has been defined. We can treat this process as one of pure 
 rational choice of the best alternative, or as a satisfying choice of one that is 
 "good enough." 
 
 These stages of decision-making are not a tidy, linear sequence of 
 steps, but are an iterative, cyclical process in which our understanding of 
 the situation, the alternatives we are considering, and our evaluation of 
 those alternatives interact with each other over time. The cyclic, iterative 
 nature of decision-making is depicted in figure 4. 
 
MIS TUTORIAL 
 
 75 
 
 Fig. 4. Diagram of the Cycles in the Decision-making Process 
 
 Decision-making at each stage and level discussed earlier can be 
 further characterized by the degree of "structure" they display. Structure 
 refers to the relative ease with which we understand and accomplish the 
 decision-making phases, and ranges from very well-structured decisions to 
 very ill-structured ones. Well-structured decisions are those that can be 
 fully specified such that a procedure can be designed to automate the 
 decision-making process. Ill-structured decisions are those that remain 
 incompletely specified and are ultimately dependent on human judgment'. 
 The basis for making those judgments rest on tacit understandings that are 
 never fully explicated. 
 
 Recently there has been increased attention on the development of 
 decision support systems. These systems use database, graphics, telecom- 
 munications, and simulation models to help managers make semistruc- 
 tured decisions where neither pure procedure, nor pure judgment prevail. 
 The emphasis is on supporting judgment by supplementing the managers 
 decision process with computer power in a way that is understandable and 
 controllable by them. 
 
 So far we have introduced the notion of a system with its nested set of 
 input, process and output relations, and we have surveyed the process of 
 management decision-making. Now, we add the perspective of the organi- 
 zation as a system. For this we use the systems approach of C. West 
 Churchman. 1 We understand organizations with a systems approach when 
 
16 RICHARD J. BOLAND, JR. 
 
 we think of the organization with five basic considerations. They are: 
 (1) the resources of the system; (2) the environment of the system; (3) the 
 components of the systems; (4) the objective of the system; and (5) the man- 
 agement of the system. 
 
 Resources are everything the system can draw you in carrying out its 
 activities. This includes everything from cash and fixed assets to depend- 
 able procedures, to employee morale and client goodwill. Managers often 
 overlook potential resources and fail to take full advantage of their 
 possibilities. 
 
 The environment includes everything that is outside the system and 
 thereby outside of its control that impacts the performance of the system. 
 For the systems approach, defining the environment correctly and adapt- 
 ing to it successfully is the critical managerial function. 
 
 The components of the organization are its missions and functional 
 programs; that which its procedures accomplish. These production pro- 
 cesses of the organization may coincide with a departmental structure, but 
 usually they will cross departmental lines, and are best conceived of as 
 organization-wide programs rather than activities of isolated subunits. 
 
 The objectives of the organization are the goals it tends to achieve. 
 These goals are contained in the recognized measures of accomplishment, 
 the criteria used for performance evaluation, and the organization's defini- 
 tion of purpose. One must be careful to distinguish the "real" from the 
 merely espoused objectives, and to observe how the organization actually 
 performs when characterizing its operating objectives. 
 
 The management of the organization is the responsible action taken 
 by its decision-makers. Here we emphasize the manager's involvement in 
 planning and control decisions. Planning decisions set standards, goals 
 and criteria over a future time horizon, and control is a process of compar- 
 ing actual achievements with planned outcomes and taking corrective 
 action as needed. This is a cybernetic feedback control process in which a 
 standard of performance is established, and results are compared to the 
 standard, prompting a managerical response when necessary. 
 
 Application of the Systems View to Libraries 
 
 This section presents some images of the way the systems approach 
 can be used to observe and understand the purposeful activity of organiza- 
 tions. The nested cyclic, input/output transformations that characterize 
 the systems approach, as well as the cybernetic control process, lend 
 themselves well to the use of visual imagery. 
 
 The intent is that these visual images be used by the manager or 
 systems analyst as a basis for exploring the set of relationships that consti- 
 tutes the system, as well as a basis for generating other images and empha- 
 
MIS TUTORIAL 
 
 17 
 
 sizing other relationships. In any event, the images are convenient ways of 
 organizing the systems concepts identified above, and applying them to a 
 library setting. 
 
 Figure 5 presents a visual depiction of Churchman's systems approach. 
 It emphasizes the organization's relationship with critical factors in the 
 environment and the demand those relationships put on the system. As 
 elements are changed, the relationships that are emphasized are changed. 
 Also, any feature of the diagram (resources, access, acquisitions) can be 
 further elaborated for detailed exploration. For instance, the management 
 planning and control process can be expanded as indicated in figure 6. 
 
 1 
 1 
 
 Environment 
 
 1 
 1 
 
 Library System 
 
 1 
 1 
 
 f 
 
 1 
 
 1 
 
 1 
 
 1 
 
 Management 
 Planning and 
 Control 
 
 1 
 
 1 
 1 
 
 1 Style 
 | Climate 
 Power 
 
 I 
 
 Objectives 
 sl' 
 
 1 
 
 1 
 1 
 
 1 
 
 
 
 1 
 Funding >. Resources ^ 
 
 Missions, 
 
 1 
 
 f Acquisitions puhHraMnna 
 
 Sources 
 
 Procedures, 
 Communication 
 
 1 
 
 I 
 
 
 1 
 
 
 Access 
 
 Fig. 5. Diagram of Churchman's Systems Approach 
 
18 
 
 RICHARD ]. BOLAND, JR. 
 
 Management Planning and Control 
 
 Strategies 
 
 Tactics 
 
 Operations 
 
 Intelligence 
 
 Design 
 
 Choice 
 
 Budgets, Standards, 
 Feedback, Rewards 
 Evaluation 
 
 Fig. 6. Management Planning and Control Process 
 
 At any level of detail that we wish to expand the diagram, we should 
 not only be concerned with identifying the missing details, but with 
 assessing the overall balance of the system, and using that assessment as a 
 basis for setting priorities. Focusing on one environmental relation while 
 ignoring others, or emphasizing one aspect of the decision-making process 
 over others is usually self-defeating. 
 
 The systems view also emphasizes the cyclic character of organiza- 
 tional processes. Figure 7 is a depiction of the library as a two-cycle system 
 of serving clients and building a collection interacting with an environ- 
 ment of knowledge creation and of publications. This basic model can also 
 be expanded, as in figure 8, to reveal subsystems in each cycle and their 
 interrelations. The process of expansion and exploration can continue as 
 each subsystem itself is depicted as a cycle of interrelated activities. Once 
 again, the benefit of this type of analysis is to assist in identifying critical 
 activities and their interrelationships, unnecessary redundancy, weakness, 
 or overemphasis. In short, it helps to explore the question of balance 
 among the many competing demands placed on the organization. 
 
 The final image presented in this section is that of a cybernetic control 
 process. For each activity in the system-in-environment diagram or in the 
 system-cycles diagrams, a control process is implied. The basic elements of 
 that control process are shown in figure 9. The model starts with the 
 familiar input-process-output diagram. Added to it is a monitor that 
 measures system outputs. The outputs are then compared to a standard, 
 goal or norm. Here the standard is shown as the prediction from a model of 
 desired system functioning. If the comparison reveals a difference, an error 
 message is received by the manager, who activates a change in the system, 
 the inputs or the standard. 
 
MIS TUTORIAL 
 
 19 
 
 Knowledge 
 Creation 
 
 Publications/ 
 Distribution 
 
 Fig. 7. Representation of the Library as a Two-cycle System 
 
 Search 
 Strategy 
 
 Holding 
 Evaluation 
 
 Inquiry 
 
 Acquisitions 
 
 Fig. 8. The Subsystem within the Basic Two-cycle System 
 
 The word cybernetic literally means "steersman" and refers to the fact 
 that communication processes set in motion by the output of an activity 
 stimulate corrective responses that tend to bring the outputs back in con- 
 trol. Thus, the system is brought into control by the act of going out of 
 control, and is an error-driven control process. 
 
 The cybernetic control model helps us explore the existence or ade- 
 quacy of the measures of system output, the feedback communication 
 channels, the model of desired system functioning, and the ability of a 
 decision-maker to take corrective action in a time frame that allows the 
 system to remain stable. If the response is too fast or too slow, the system 
 will display oscillations around the standard, but will not converge on it. 
 
 We generally recognize three orders of cybernetic feedback control. 
 First order feedback returns system outputs to an acceptable range, given a 
 standard goal. Second order feedback modifies the goal itself to maintain 
 
20 
 
 RICHARD J. BOLAND, JR. 
 
 Model 
 
 Prediction 
 
 Decision 
 Maker 
 
 Comparison 
 
 son 
 ~~ 
 
 Organizational 
 
 Resources 
 and Components 
 
 Monitor 
 
 Fig. 9. Diagram of the Basic Elements in the Control Process 
 
 an overall strategy in a changing environment, and third order feedback 
 modifies the strategy and purpose of a system based on a learning process 
 that questions the nature of the systems relation with its environment. 
 
 Implications for Deriving Management Information from Automation 
 
 We now add to the discussion above some further considerations in 
 designing management information systems. 
 
 The organization itself the way it is structured with routine proce- 
 dures and reporting relations is a source of management information. 
 These in-place procedures not only provide the positive or negative feed- 
 back that enables cybernetic control, they also define the organization's 
 ability to sense changes in the environment, react to disturbances, handle 
 exceptions, implement plans, achieve consensus, and revise plans. The 
 structure and process of the organization thus deserves as much attention 
 as the data processing support available to a particular manager. 
 
 Another consideration is that the data system must fit the organiza- 
 tion. The style, climate and power that characterizes the management of an 
 organization is critical to effective information system design. Is the struc- 
 ture based on type of client served, library function, geographic location, 
 or a matrix combination? Is the structure centralized or decentralized, 
 formal or informal? Is there agreement on goals? How are unit heads held 
 accountable? What kinds of reward and status systems are in place? How 
 
MIS TUTORIAL 21 
 
 freely does formal and informal communication occur? What style of 
 leadership does the organization display rational persuasion, inspira- 
 tion or empathy? How are these tied to the methods of motivating 
 employees? What degree of formality, concern for the privileges of office, 
 responsibility for worker security, and participation in decision-making 
 characterize the organization? Where does power rest in the organization? 
 How is it displayed and used? All these considerations are important 
 because they contribute to the crucial difference between data and informa- 
 tion. The data from an automated system will only be informative in an 
 organizational context. To realize its potential value, data must lead to 
 understanding and effective action, both of which are constrained by the 
 organization structure. 
 
 We observe that most organizations have a limited number of critical 
 factors that spell the difference between success and mere survival for an 
 organization. This limited number of critical factors follows from the 
 organization's strategic relations with its environment. Key decisions 
 related to these critical factors is where the effort of management informa- 
 tion support should be directed. There is practically no limit to the number 
 of decisions and management activities that could be identified and sup- 
 ported with automation. Most of them are not worth the effort. The value 
 of system development is maximized when those key decisions that affect 
 the critical success factors receive the focus of attention and effort. 
 
 Another consideration is that a cybernetic control image emphasizes 
 the importance of standards of evaluation and models of system perfor- 
 mance. Unless there are standards to which actual outcomes a re com pa red, 
 there can be no stimulus for corrective action. Unless there is at least an 
 implicit model of how the organization should be functioning and how 
 decisions should be made, there is no basis for learning. The definition of 
 standards for evaluation and the identification of the decision models 
 managers do or should use is perhaps the most significant outcome of 
 developing a management information system. 
 
 Data to provide management information may be generated from: 
 
 1. Reports from operations and transaction-based systems. The reports 
 can be regularly scheduled, ad hoc, or exception based, with content tied 
 to the level and type of decision being supported. 
 
 2. Access to database systems, both internal and external, that allow in- 
 quiry and special reports. 
 
 3. Modeling facilities that allow simulations, statistical analyses and 
 forecasts. 
 
 In addition to database access and statistical and graphic analysis, manage- 
 ment terminals can also offer time management, project management, 
 message management, and teleconferencing services. 
 
22 RICHARD J. BO LAND, JR. 
 
 The types of decision support models that can be developed include: 
 
 1. probabilistic decision models, where the alternatives and payoffs identi- 
 fied by the decision-maker are combined with their expectations of the 
 occurrence of future events; 
 
 2. deterministic simulations showing how a closed-system set of relation- 
 ships behave over time. Cash flow and budget projections are classic 
 examples of these simulations, and are the basis of financial planning 
 and control; 
 
 3. forecasting models where historical experience of demand, usage, etc. is 
 extrapolated to generate data for capacity planning and other purposes; 
 and 
 
 4. optimizing models, such as linear programming, where a set of con- 
 straints are taken into account in maximizing an objective function. 
 
 Perhaps the most common type of managerial support, however, is 
 provided by a set of search, sort and statistical programs tied to a large file 
 representing a portfolio of objects for which the manager has responsibil- 
 ity. For instance, we have recently been involved in developing an acquisi- 
 tion support system for a media center. It consists of a file of potential 
 acquisitions along with a boolean search procedure, sort and statistical 
 procedures, and a report generator. In the development of the system, there 
 were many critical technical issues that needed careful attention and 
 which, in some instances, constrained the design. Yet, the most crucial 
 issues in design centered on how the system was going to change the 
 location of decision-making on acquisitions, with some people losing 
 power and others gaining it. 
 
 Although the software of this system is powerful and flexible, its 
 effectiveness will depend on the quality with which managers rank target 
 areas and rate potential acquisitions. This kind of formal quantification is 
 a new behavior that must be learned. Finally, the boolean search proce- 
 dures are only as good as the questions the decision-makers will ask. The 
 system only presents a potential, and the manager alone has the possibility 
 to realize it which leads to the final implication. 
 
 The major reason for the failure of management support systems are 
 organizational not technical or economic. To be a success, the system 
 must be implemented and used. This is a question of organizational 
 acceptance and individual learning on the part of managers. There is no 
 clear recipe for implementation success, but some prominent features of 
 systems that succeed are: 
 
 1. that there is a strong felt need on the part of managers to develop the 
 system; 
 
 2. that top administration personnel supports and fosters the effort; 
 
MIS TUTORIAL 
 
 23 
 
 3. that all affected parties are actively, meaningfully involved in the devel- 
 opment; and 
 
 4. the system is congruent with the climate, style and power of the organi- 
 zation. 
 
 The System Development Process 
 
 This section gives a brief overview of an ideal system development 
 process. The design, installation and evaluation of management support 
 systems is a tightly woven cycle that displays an evolutionary, adaptive 
 learning capacity. It is an iterative, recursive process that is easily separated 
 into neat stages only in papers such as this. Figure 10 depicts the system 
 development process as beginning with a system plan. My intention is to 
 highlight the need to identify the critical decisions needing support based 
 on the organization's strategic relations with its environment. The issues 
 of efficiency versus effectiveness discussed in section one must be resolved 
 in the planning process with a time-phased identification of priorities. 
 The plan should chart the organization's forward movement by maintain- 
 ing a balance among its key functions, and its level of managerial and 
 technical sophistication. 
 
 System 
 Planning 
 
 Review and 
 Evaluation 
 
 ^ Feasibility 
 Study 
 
 User 
 Experience 
 
 Operation and 
 Maintenance 
 
 Analysis and 
 Design 
 
 Fig. 10. Diagram of the System Development Process 
 
24 RICHARD J. BOLAND, JR. 
 
 The planning process, combined with the experience of user groups, 
 creates a felt need for change and system development. It is this felt need for 
 change which should drive system design. The feasibility study is an 
 opportunity to assess the quality of this need and to test the economic, 
 technical and operational validity of the proposed development. It is 
 important here to focus on the decisions that will be supported, their 
 significance and the impact the system will have on them. This requires 
 that the study team understand the decision process in question, and that 
 they do not merely assume that automation will enhance it. 
 
 The proposal should be assessed in terms of its fit with the style, 
 climate and power of the organization and with the openness of the parties 
 involved to accept a change. This requires that there be a dissatisfaction 
 with the existing state of affairs and a willingness to experiment and learn 
 new behaviors as a group, as well as at an individual level. 
 
 Analysis and design requires active involvement and support between 
 the manager and the system analyst or technical experts. The best form of 
 this involvement combines a sharing of design responsibility with a sense 
 of mutual understanding, in which each participant respects and attempts 
 to understand the perspective and concerns of the other. This type of 
 involvement requires a significant time commitment by managers, and if 
 they are not prepared to give it, perhaps the felt need is not as great as was at 
 first thought. 
 
 The analysis and design stage can follow a top-down or a bottom-up 
 progression. Top-down entails movement from goals and objectives to a 
 logical system that meets their requirements, while bottom-up entails 
 starting with existing procedures and processes and designing an 
 improved system. Usually both used in conjunction with each other will 
 prove most effective. This is because the decision process is not just a 
 rational process of selecting best courses of action. 
 
 As mentioned above, the procedures, programs, offices, and routines 
 are an important, organizationally-embedded source of decision-making 
 and action. In fact, I would argue that most organizational decisions are 
 determined by the interaction of routine organizational procedures. The 
 decision process is also a political, disjointed one in which coalitions form 
 and dissolve as threats and opportunities change. Any analysis that 
 emphasizes the purely rational at the expense of appreciating the proce- 
 dural and political is risking implementation failure. 
 
 An evaluation of the design should be performed before programming 
 and testing. The risk of implementation failure should be reassessed, as 
 well as the value of the system to the intended decisions. Designs have a way 
 of being modified over time and this provides a test that the expected 
 impact on crucial decisions has not been lost. 
 
MIS TUTORIAL 25 
 
 Installation, operation and maintenance are beyond the scope of this 
 paper, except to say that structured approaches to design and program- 
 ming seem to produce systems with fewer errors and less delays. Also, 
 systems designed with structured techniques appear to require less effort 
 for maintenance and modification. Because of the significant percentage of 
 effort consumed by routine maintenance, this can have a major impact on 
 the resources available for the development of new and improved systems. 
 
 Review and evaluation of systems after implementation is one step in 
 the development process that is frequently ignored. Yet the periodic review 
 of existing systems is necessary to weed out those that are no longer cost 
 effective. Industrial firms that have started determined efforts at postimple- 
 mentation reviews find many systems that are overly-sophisticated. These 
 firms are freeing up computer resources by shifting these applications 
 away from online and toward batch processing. 
 
 Perhaps most importantly, an effort at postimplementation review 
 focuses management attention on the all important questions of assessing 
 the efficiency and effectiveness of their organization. It requires defining 
 desirable outcomes that can serve as a basis for measuring the quantity and 
 quality of their output. Ultimately, this assessment of the efficiency and 
 effectiveness of outputs, and the identification of decisions critical to their 
 improvement is the driving force behind the entire system development 
 process. 
 
 Conclusion 
 
 This paper has briefly covered a great deal of ground in surveying a 
 major part of the management information system area. While it could not 
 cover each area in the depth it deserves, the hope is to provide a framework 
 for management information support within which the other, more 
 detailed papers can find a common ground. 
 
 REFERENCE 
 
 1. Churchman. C. West. The Systems Approach, 2d ed. New York: Basic Books, 1979. 
 
 ADDITIONAL REFERENCES 
 
 Emery, James C. Organizational Planning and Control Systems. New York: Macmillan, 1969. 
 Keen, Peter CW., and Scott Morton, Michael S. Decision Support Systems. Reading: 
 Addison- Wesley, 1978. 
 
26 RICHARD J. BOLAND, JR. 
 
 Schoderbek, Charles G. Management Systems: Conceptual Conserations, 2d ed. Dallas, 
 
 Tex.: Business Publications, 1980. 
 Simon, Herbert A. The New Science of Management Decision, rev. ed. Englewood Cliffs, 
 
 N.J.: Prentice-Hall, 1977. 
 Tricker, Robert I., and Boland, R.J. Management Information and Control Systems, 2d ed. 
 
 London: Wiley, Chichesler, 1982. 
 
KEN DOWLIN 
 
 Director 
 Pikes Peak Library District 
 
 LYNN MAGRATH 
 
 Associate Director 
 Pikes Peak Library District 
 
 Beyond the Numbers 
 A Decision Support System 
 
 The degree of sophistication of an organization's information system is 
 indicated by its placement in the stages of evolution of general information 
 systems. The first stage is the automation of clerical tasks. The second stage 
 is redesign of the system and subsystem integration. The third stage is 
 support for middle management decisions, and the fourth stage is support 
 for top level, decision-making. 1 The library community has progressed 
 through the stage of automating clerical functions. There are computer- 
 ized systems for almost all such systems in libraries. These systems are not 
 in operation in every library, nor are all of the systems in operation in any 
 single library but the pieces are there. Libraries are just starting into the 
 second stage systems redesign and integration. There are a handful of 
 libraries that have implemented major portions of an integrated system. 
 There are even fewer libraries that have reached the third stage and almost 
 no library that has reached the fourth stage. 
 
 The Pikes Peak Library District (PPLD) completed the first stage in 
 the development of its general information system in 1979, the majority of 
 the second and third stages at this writing, and has embarked on a serious 
 effort to complete the fourth stage. 
 
 Information for top level decision-making must, by necessity, include 
 data from sources that cannot be automated. It is very difficult to include 
 information about the environment on the library's computer in a way that 
 the computer can collect that data in a routine fashion. The process of 
 coping with changes in the external environment may be the most difficult 
 part of the planning process, and may be the most crucial. 2 All of the 
 routine data about the operations of the organization which are automated 
 may be collected and stored by the computer. The computer may even 
 organize the data in a way to assist its interpretation. And though the 
 
 27 
 
28 KEN DO WUN b L YNN MA GRA TH 
 
 computer allows us to collect a large mass of data easily, the task that 
 remains is to design the systems that provide the framework for this data, 
 the analysis of the data, and the criteria which the computer may use to 
 identify exceptions. 
 
 While designing the computer system is an important component of 
 the decision support system (DSS), a recent study by Cheney and Dickson 
 supports the proposition "that it doesn't make a great deal of difference 
 what computer technology an organization employs.. ..The important 
 thing is how it is used. " The results of their study, published in Academy of 
 Management Journal, indicated (1) how user interaction takes place, 
 
 (2) how employees are managed, and (3) how computer resources are allo- 
 cated within the organization. All are more important in terms of user satis- 
 faction and use of the system than what hardware or software is employed. 3 
 
 Robert N. Anthony provides a useful framework for categorizing the 
 collected data into levels. He suggests that the information be assigned to 
 three classes: (1) strategic planning, (2) management control, and 
 
 (3) operational control. Strategic planning is defined as the process of de- 
 ciding on objectives of the organization, on changes in these objectives, on 
 the resources used to attain these objectives, and on the policies that are to 
 govern the acquisition, use and disposition of these resources. In the public 
 library, strategic planning is the responsibility of the board of trustees, 
 where thay have the legal authority to do so. If not, it is the responsibility of 
 the city or county council, or other elected body. The library director is also 
 involved in providing recommendations to the board for their considera- 
 tion. In academic or school libraries, strategic planning is done by the 
 entity charged with the overall responsibility for the institution, and the 
 library is usually considered an entity for the implementing of total orga- 
 nization goals. 
 
 Management control is the process by which managers ensure that 
 resources are obtained and used effectively and efficiently in the accom- 
 plishment of the organization's objectives. In the library world, this is 
 usually the task of the library administration. 
 
 Operational control is the process of assuring that specific tasks are 
 carried out effectively and efficiently. This level is the responsibility of the 
 department or division supervisors. 4 
 
 A DSS is not a single invention or method. It is a new direction or 
 philosophy in computer-supported management. It builds a database to 
 answer some question or to work on a problem as yet undefined or 
 unstructured. 6 Such a system also provides the ability to manipulate the 
 data to answer those questions that have not been formulated at the time 
 that the system is designed. The payoff for a DSS is its ability to extend the 
 range and capability of managers to make decisions without imposing 
 solutions. 7 
 
DECISION SUPPORT SYSTEM 29 
 
 "The old MIS [management information system] design approach 
 begins with existing systems, and produces benefits by chance. The 
 approach that [should be followed] focuses on key tasks and decisions 
 leading to more effective decisions, and then attacks the problem of design- 
 ing information systems to support those tasks." 8 The following general 
 model may be adopted as a guideline for a DSS adaptable to the library 
 setting. This model holds that: 
 
 1. Data categories should be mutually exclusive. 
 
 2. Data should be capable of aggregation. 
 
 3. Data should be capable of disaggregation to very specific levels. 
 
 4. The data systems should accommodate complex organizations and 
 small organizations equally well. 
 
 5. Data categories should be understandable by the parent organization 
 and lay persons, as well as librarians and library staff. 
 
 6. Data should be organized into a framework intelligible to both lay 
 and library persons. 
 
 The DSS for the Pikes Peak Library is the result of an amalgamation 
 of various factors which include the data that has been historically col- 
 lected; adding methodologies to aggregate data; designing a framework for 
 the total system; using the report writer on "Maggie" (PPLD's computer) 
 to store, organize, and retrieve the data; adding mathematical formulas 
 that provide indicators for measurement; and including new data that had 
 relevance. It has been in development for over five years, since one of the 
 original objectives of the automation of the library operations was to 
 provide management information. As the automation of the library's 
 functions became reality, significant elements were added to the DSS. A 
 look at the overall scope of "Maggie's Place," the automated environment 
 of the Pikes Peak Library, is contained in Ken Dowlin's book titled The 
 Promise of the Electronic Library. 
 
 The functions of the DSS are: 
 
 1. To generate the monthly and annual reports required by external 
 entitities (accountability). Current data generated for these accountability 
 reports is often compared with historical trends and provides support for 
 ongoing decision-making by managers and supervisors. The resulting 
 decisions will directly affect the operation of the library on a short- and 
 medium-range basis. This information can be a valuable tool in making 
 decisions such as which programs should be changed or eliminated. 
 Because many of these same accountability statistics are kept by most 
 libraries they provide a point of comparison between libraries as well as a 
 device for internal comparison and external reporting. 
 
 2. To provide structured data that can be manipulated to provide 
 analysis on a scheduled basis. "Scheduled analysis" is generally performed 
 
30 KEN DO WLIN b LYNN MA GRA TH 
 
 to evaluate a specific function longitudinally. This type of analysis may be 
 carried on indefinitely in order to provide continuous input for assessing 
 the performance of a particular variable. Two examples are, the inventory 
 use ratio (IUR) and the patron use ratio (PUR). These two measures are 
 routinely generated by PPLD's computer in response to specific informa- 
 tion needs which were not met by the traditional level one accountability 
 measures. The IUR was created to confirm the validity of the materials 
 selection decisions made by the PPLD librarians. Once a month the IUR 
 reports the percentage of each classification that is checked out at the point 
 of data collection. Longitudinal studies using the inventory use ration will 
 confirm or deny the validity of the assumption that the IUR is an accurate 
 predictor of future use, hence a useful tool for determining future materials 
 selections. 
 
 The patron use ratio has provided some surprising information con- 
 cerning use patterns of patrons. Recent studies show that the majority of 
 library card usage by patrons is primarily from those who have obtained 
 their library cards within the last year. As this indicator has been developed 
 recently, monthly data collection over the next two years will be used to 
 validate the early findings. 
 
 3. To answer a specific question or to validate assumptions (investiga- 
 tion). Do the members of the community perceive the library as a commu- 
 nity information center, as a community communications center, and as a 
 source of books and materials for the community? Does the community 
 perceive the library as an organization which operates efficiently? These 
 are the types of questions which provide valuable input for future decision- 
 making concerning overall library goal setting and resource allocation. 
 The perceptual survey was designed to answer these specific questions. 
 This survey has shown that the library has exceeded its goal when 51 
 percent of the community perceives the library as a materials center and 
 when 51 percent of the community believes that the library operates 
 efficiently. If the board of directors continues to support these specific 
 goals, this survey will be repeated again in approximately two years to find 
 out if the library has made progress toward the accomplishment of the 
 other two goals. 
 
 4. To provide intersystem validation (validation). This level affords 
 the decision-maker the option of comparing the data gathered by the 
 specific studies conducted at level three. For example, the perceptual 
 survey contains data detailing how many times the respondents (a repre- 
 sentative sample) have used the library within the last year. This informa- 
 tion can be used to double check and validate the results found in the 
 patron tenure reports. Information regarding the sex and age of users can 
 also be validated using these two tools. 
 
DECISION SUPPORT SYSTEM 31 
 
 The initial framework for the DSS was modeled after the proposed 
 framework developed by the National Center for Higher Education Man- 
 agement. 10 Significant changes have occurred since the PPLD system was 
 documented in Library Effectiveness: A State of the Art in 1980. 11 
 
 In 1979 the library board of trustees adopted a long-range plan that 
 contained the statement of the mission of the library and its proposed 
 goals. 12 They decided that the mission of the library is (1) to serve as the 
 resource center for published materials for the community, (2) to serve as 
 the community information center, (3) to serve as the community com- 
 munication center, and (4) to operate as an efficient organization with goal 
 attainment to be planned on a consistent basis. 
 
 Goals were adopted which required that 51 percent of the community 
 should perceive the library as meeting the needs of the community for each 
 of the mission elements. In 1980 it was estimated that the library met or 
 exceeded the requirements for a resource center (51 percent or more); that, 
 as an information center, it was below the level of the goal (less than 30 
 percent); that its communications were below the level of the goal (less 
 than 10 percent); and that its efficiency was around the goal (51 percent). 
 
 In 1981, the community was surveyed to determine its perception of 
 the library. The hypotheses of this survey were: 
 
 1. That 51 percent of the community perceived the library as a materials 
 resource center for the community (predicted less than 50 percent). 
 
 2. That 51 percent of the community perceived the library as an 
 information center for the community (predicted less than 30 
 percent). 
 
 3. That 51 percent of the community perceived the library as a 
 communications center for the community (predicted less than 10 
 percent). 
 
 4. That 51 percent of the community perceived that the library operates 
 in an efficient manner (predicted less than 50 percent). 
 
 Methodology 
 
 The telephone survey was conducted by students in a graduate level 
 statistics class during fall semester 1981 . The respondents were chosen on a 
 random basis from the Colorado Springs metropolitan area telephone 
 book. Only one questionnaire per household was completed, and survey- 
 ors were instructed not to interview potential respondents under the age of 
 fourteen. A total of 214 people were questioned. If more than one answer 
 was given to a question the response was left blank by the data coder. 
 Written guidelines were provided to each surveyor including such instruc- 
 tions as a standard definition of efficient and other anticipated points of 
 
32 KEN DO WL1N b L YNN MA GRA TH 
 
 clarification the surveyor might not be automatically familiar with. The 
 questions relating directly to the goals gave the respondent three possible 
 choices: agree, disagree, or don't know. 
 
 Characteristics of Respondents 
 
 The survey respondents were between the ages of twenty-five and 
 thirty-four years of age (27 percent). The age groups thirty-five to forty- 
 four and forty-five to fifty-four were second in frequency with each cate- 
 gory containing 18 percent of the respondents. Of these, 65 percent of the 
 respondents were women. 
 
 Total Income of Respondents' Households 
 
 Of the respondents surveyed, 28 percent reported their total family 
 income to be between $10,000 to $19,000; 20 percent reported their total 
 family income to be between $20,000 to $29,999; 16 percent reported their 
 total family income to be between $0 to $9,999; and 10 percent reported the 
 total family income to be between $30,000 to $39,999. Demographic data 
 were collected whenever possible in categories corresponding to those used 
 by the U.S. Census to validate the sample. The income percentages do not 
 differ significantly from those reported by the 1980 census for the popula- 
 tion sampled. At this writing, age and educational level data are unavail- 
 able for the 1980 census of the tracts involved. 
 
 Respondents' Use of the Library in the Past Year 
 
 The telephone survey also provided information regarding respon- 
 dents' use of the library in the last year. The survey contained a good 
 representation of both users and nonusers. Of those surveyed, 33 percent 
 used the library zero times in the last year; 30 percent had used the library 
 six times or less; 25 percent had used the library approximately twelve 
 times; 9 percent had used the library approximately fifty-two times; and 3 
 percent reported use of the library more than fifty-two times in the last year. 
 
 Conclusions 
 
 Of those surveyed, 67 percent perceived the library as a place to find 
 "materials" (only 2 percent disagreed, the remaining 31 percent didn't 
 know). In addition, 39 percent of those surveyed perceived the library as a 
 place to find information (only 4 percent disagreed, the remaining 57 
 percent didn't know). Well above the predicted percentile, 34 percent 
 
DECISION SUPPORT SYSTEM 33 
 
 perceived the library as a center for communications for the community (5 
 percent disagreed, and 61 percent didn't know). 
 
 Even more surprising were the efficiency figures: 73 percent perceived 
 the library as using its resources efficiently (4 percent disagreed, and 23 per- 
 cent stated they did not know). One of the most striking points to note is 
 that so few people disagree in any of the categories, particularly "efficien- 
 cy." A table of the current DSS is helpful for determining the relationship 
 of the components to Anthony's levels for decision-making (see table 1). 
 
 Reports Produced by the Pikes Peak Library District 
 
 Compliments and Complaints. These are comments on the library 
 and its operations by the users or nonusers. They are written and sent to the 
 supervisor and manager concerned, the library director, and the board of 
 trustees. The comments are not aggregated, but they do provide a sense of 
 the operation of the library. 
 
 Budget and Expenditures. This is a monthly report of all budgeted 
 amounts and expenditures by operational unit and line item. Each super- 
 visor and manager receives the section on his area of responsibility, and an 
 aggregate is provided to the board of trustees and external agencies. This 
 data can be displayed on a terminal at any time with the data current to the 
 minute. A sample report is contained in figure 2. 
 
 Revenues. This monthly report indicates the revenue received and 
 compares it to previous time periods. Since the PPLD is a semi- 
 autonomous agency, expenditures cannot exceed revenues. Since the 
 majority of the budget is based on estimates from the county assessor, care 
 must be taken to monitor cash in the event the estimates are not accurate. 
 This data is available online to the financial development office and the 
 director. The revenue report is illustrated in figure 3. 
 
 Service Indicators. These totals are provided monthly to the board, 
 director and managers. The data is loaded into the design support system 
 by the division that is collecting them or the computer from operational 
 programs. This report and the historical data can be accessed and analyzed 
 online. Figure 4 provides a sample report. Figure 4a shows the codes and 
 definitions for the items. The items are grouped under the goal to relate to 
 goal attainment. 
 
 Inventory Count. This count is a historical continuation of the long 
 standing tradition of counting items added and deleted. A complete inven- 
 tory of the collection using the computer has just been finished. When this 
 inventory is validated, the figures in the computer will displace those of the 
 manual system. A sample report is contained in figure 5. 
 
TABLE 1 
 
 CURRENT DECISION SUPPORT SYSTEMS FOR 
 PIKES PEAK LIBRARY DISTRICT 
 
 Board of Trustees 
 (strategic planning) 
 
 Compliments and complaints (m) 
 
 Budget and expenditure reports (m) 
 
 Revenue reports (m) 
 
 Service indicators reports (m) 
 
 Inventory counts (m) 
 
 Perceptual surveys (nr) 
 
 Project status reports (m) (nr) 
 
 Director 
 (strategic planning and management control) 
 
 Inventory use ratio exceptions (m) (o) 
 
 Inventory use ratios (m) (o) 
 
 Budget and expenditures by department and division (m) (o) 
 
 Flow through expenditures and revenues by account (m) 
 
 Patron use (m) (o) 
 
 Patron tenure (nr) 
 
 Perceptual survey (nr) 
 
 Security reports (d) 
 
 Online catalog use counts (d) (m) 
 
 Compliments and complaints (d) 
 
 Critical incidents (d) 
 
 Item information (o) 
 
 Online catalog survey (nr) 
 
 Project status reports (m) (nr) 
 
 Service indicators reports (m) (o) 
 
 Revenue reports (m) (o) 
 
 Managers 
 (management control and operations control) 
 
 Public Services Department 
 
 Service indicators report (m) (o) 
 
 Budget and expenditures report (m) (o) 
 
 Materials encumbrances (m) 
 
 Inventory use ratio (m) (o) 
 
 Inventory use ratio exceptions (m) (o) 
 
 Patron use (m) (o) 
 
 Information and reference detail report (m) (o) 
 
 Catalog use counts (d) 
 
 Compliments and complaints (d) 
 
 Perceptual survey (nr) 
 
 Critical incidents (d) 
 
 Online catalog survey (nr) 
 
 Periodicals use report (t) 
 
 Item information (o) 
 
DECISION SUPPOR T SYSTEM 35 
 
 TABLE \-Continued 
 
 Technical Services Department 
 Inventory counts (m) 
 Critical incidents (m) 
 Challenged materials (d) 
 Item information (o) 
 Material encumbrances (m) 
 Budget and expenditure reports (m) (o) 
 
 Operations Department 
 
 Service indicators (m) (o) 
 Circulation detail report (m) 
 Security reports (d) 
 Critical incidents (d) 
 Patron information (o) 
 Item information (o) 
 
 Ridefinders (project) 
 
 Service indicators (m) (o) 
 Budget and expenditures (m) (o) 
 Project status reports (m) 
 
 o Data can be accessed and analyzed online 
 
 nr Data is collected only when required 
 
 d Daily reports 
 
 m Monthly reports 
 
 t Data is collected three times per year 
 
 Perceptual Survey. This survey provides information concerning the 
 community's perceptions toward the library in relation to the library's 
 mission. Figure 6 shows the perceptual survey data. 
 
 Project Status Reports. These reports are generated by special projects 
 such as the "Past as Prologue Series," funded by the National Endowment 
 for the Humanities, and Ridefinders, which is funded by the Pikes Peak 
 Area Council of Governments and operated by the library. 
 
 Inventory Use Ratio Exceptions. This report is created from the 
 inventory use ratio program. After the inventory use ratio is calculated, the 
 program calculates the mean for all the classes, the standard deviation, and 
 identifies the classes that are over one standard deviation above or below 
 the mean. These classes are identified as exceptions allowing investigation 
 into each specific class. A sample report is shown in figure 7. 
 
 Inventory Use Ratio. When commanded to do so, the computer reads 
 the entire inventory file and produces the inventory use ratio report. This 
 run is usually done on the last Sunday of each month. The computer 
 counts the number of items in each class, counts the number of items 
 checked out, and the number of items that are noncirculating. The inven- 
 tory use ratio is calculated by dividing the number of items checked out by 
 
Executive Council 
 
 Public Services, Lynn Magrath 
 
 Technical Services, Brenda Hawley 
 
 Operations , Jim Graham 
 
 Financial Dev. , Dorothy Sausser 
 
 Human Res., Anne Fitzgerald 
 
 Systems, Rick Richmond 
 
 Public Info., Nancy De Lury 
 
 Transpool, Judy Evans 
 
 Management Team 
 
 Fig. 1. Organizational Structure of the Pikes Peak Library District (PPLD) 
 
SALARIES 
 HOURLY SALARIES 
 TEMPORARY HOURLY 
 PARTIAL PAYMENT 
 PAYROLL CONTRIBUTIONS 
 TRAINING 
 TOTAL PERSONNEL 
 
 Budgeted 
 
 $1,237,388.00 
 60,862.00 
 7U.721.00 
 8.79U.OO 
 233,777.50 
 9,270.00 
 $1,62U,812.50 
 
 Expended 
 To Date 
 
 $305,97U.l8 
 13.782.36 
 23,923.33 
 877. U5 
 5U.570.05 
 2.U26.17 
 $U01,553.5U 
 
 Expended 
 This Mo. 
 
 $10U,237-88 
 U.U39-90 
 8.756.9U 
 1.7U2.23 
 17,105-03 
 1,397-79 
 $13U,195.31 
 
 $ 
 $1 
 
 Current 
 Balance 
 
 931.U13-82 
 U7.079-6U 
 50,797-67 
 7,916-55 
 179, 207- U5 
 6.8U3-83 
 ,223.258.96 
 
 LEASED BOOKS 
 BOOKS 
 MICROFORMS 
 SUBSCRIPTIONS 
 STANDING ORDERS 
 PREVIEW MATERIALS 
 FILMS & FIIK3TRIPS 
 BINDING 
 RECORDINGS 
 FRAMED PRINTS 
 TOTAL MATERIALS 
 
 $ 63.22U.UU 
 128.8lO.l6 
 27,358.90 
 31.U83.00 
 2U.o66.00 
 U5.U6o.00 
 8,350.00 
 10,000.00 
 2,150.00 
 0.00 
 $ 3U0.902.50 
 
 $ 6,715.80 
 15.26U.17 
 2.277.65 
 27.890.U8 
 2,753.75 
 13.782.U5 
 593-09 
 3,183. U6 
 UU.80 
 0.00 
 $ 72,505.65 
 
 $ 1,755.00 
 9,919.60 
 0.00 
 830. Ul 
 1.U03-79 
 7,176.61 
 0.00 
 1,880.87 
 0.00 
 0.00 
 $ 22,966.28 
 
 $ 
 $ 
 
 56.508.6U 
 113.5U5-99 
 25.08l.25 
 3,592.52 
 21,312.25 
 31,677.55 
 7,756.91 
 6.816.5U 
 2,105.20 
 0.00 
 268,396.85 
 
 GENERAL SUPPLIES 
 SPECIAL SUPPLIES 
 TOTAL SUPPLIES 
 
 $ 26.9U5.00 
 90,271.63 
 $ 117,216.63 
 
 $ 3,916.05 
 26.9l8.lU 
 $ 30.83U.19 
 
 $ 361.30 
 U, 956. 55 
 * 5,317.85 
 
 $ 
 $ 
 
 23,028.95 
 63.353.U9 
 86.382.UU 
 
 COMPUTER OPERATION 
 TELEPHONE 
 UTILITIES 
 CONTRACT SERVICES 
 FACILITIES RENTAL 
 VEHICLE MAINTENANCE 
 INSURANCE 
 TOTAL FACILITIES 
 
 $ 80,Ul2.00 
 32,678.00 
 76.6Ul.00 
 lU2.U22.73 
 57,675.00 
 13.OUo.00 
 10,803.00 
 * U13.671-73 
 
 $ 21,397.28 
 7,925.79 
 15.UU8.38 
 3U.829.89 
 18,526.50 
 2,690.72 
 237-00 
 $101,055,56 
 
 I 2.29U.08 
 2,956-UU 
 10,217.27 
 13,627.65 
 U, 681. 00 
 2,361.98 
 237-00 
 $ 36.375-U2 
 
 $ 
 $ 
 
 59.OlU.72 
 2U.752.21 
 61,192.62 
 107, 592. 8U 
 39.1U8. 50 
 10.3U9.28 
 10,566.00 
 312.6l6.17 
 
 BUILDING REPAIR 
 FURNITURE 
 VEHICLES 
 EQUIPMENT 
 TOTAL CAPITAL ITEMS 
 
 $ Ul.200.00 
 3,706.00 
 U, 300. 00 
 
 26,235.72 
 
 $ 75.UUl.72 
 
 $ 16,757. U9 
 609.00 
 0.00 
 12.15U.UO 
 $ 29,520.89 
 
 $ 730.77 
 U36.50 
 0.00 
 3,319-99 
 $ U.U87-26 
 
 $ 
 $ 
 
 2U.UU2.51 
 3,097.00 
 U, 300. 00 
 1U,081.32 
 U5.920.83 
 
 INSTITUTIONAL DUES 
 TREAS./ AUDITOR FEES 
 RESEARCH & DEVELOPMENT 
 MILEAGE & TRAVEL 
 TOTAL MISCELLANEOUS 
 
 $ 2,100.00 
 33,000.00 
 
 170,286.00 
 
 12,511.00 
 $ 217,897.00 
 
 $ 1,085.00 
 U.7U0.09 
 U, 223. 27 
 1,921.37 
 $ 11,969.73 
 
 $ 0.00 
 3.822.0U 
 1,023-Ul 
 U35-50 
 $ U.U09-95 
 
 $ 
 $ 
 
 1,015-00 
 28,259.91 
 166,062.73 
 10.589-63 
 205,927-27 
 
 TOTAL BUDGET 
 
 $2,789,9U2.08 
 
 $6U7,U39.56 
 
 $207,752.07 
 
 $2 
 
 .lU2.502.52 
 
 Fig. 2. Sample Budget Report for the PPLD 
 

 
 O 
 
 o 
 
 O 
 
 O 
 
 
 
 O 
 
 O 
 
 o 
 
 
 
 O 
 
 o 
 
 O 
 
 
 
 o 
 
 O 
 
 O 
 
 o 
 
 
 cu 
 
 _^ 
 
 UN 
 
 O 
 
 O 
 
 o 
 
 H 
 
 CM 
 
 00 
 
 
 -p 
 
 c~- 
 
 f- 
 
 
 
 o 
 
 o 
 
 OJ 
 
 CM 
 
 J- 
 
 
 1 
 
 oo 
 
 t 
 
 o 
 
 o 
 
 o 
 
 CO 
 
 H 
 
 ON 
 
 
 H 
 H 
 
 o 
 
 OJ 
 
 o 
 
 LPv 
 
 o 
 
 H 
 
 CO 
 
 ON 
 
 
 -p 
 
 o 
 
 H 
 
 
 00 
 
 CM 
 
 CO 
 
 OO 
 
 H 
 
 
 CO 
 
 H 
 
 
 H 
 
 
 
 OO 
 
 OO 
 
 t 
 
 
 W 
 
 fk 
 
 
 
 
 
 
 ft 
 
 n 
 
 
 
 CM 
 
 
 
 
 
 
 CM 
 
 CM 
 
 
 Tj 
 
 
 
 
 
 
 
 
 
 
 -P 
 
 fe^. 
 
 ^5. 
 
 ^S. 
 
 fes. 
 
 Vi 
 
 tsS. 
 
 &. 
 
 ^. 
 
 
 O 
 "fe^. CU 
 
 -* 
 
 CM 
 
 OO 
 
 CM 
 
 OJ 
 
 00 
 
 * 
 
 t- 
 
 
 H 
 
 H 
 
 ON 
 
 OO 
 
 OO 
 
 H 
 
 t 
 
 CM 
 
 H 
 
 
 H 
 
 CM 
 
 UN 
 
 00 
 
 OO 
 
 
 H 
 
 CM 
 
 CM 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 
 JH -^ 
 S3 
 
 CO 
 H 
 
 
 CM 
 OJ 
 
 00 
 ON 
 
 
 CO 
 CO 
 
 OO 
 OO 
 
 H 
 CM 
 
 
 <U Q 
 
 t 
 
 
 OJ 
 
 t 
 
 
 t~- 
 
 [ 
 
 UN 
 
 
 S 
 
 t 
 
 
 
 CM 
 
 
 ON 
 
 t- 
 
 t 
 
 
 
 CM 
 
 
 CO 
 
 CO 
 
 
 
 o\ 
 
 VO 
 
 
 " W 
 
 fl 
 
 
 fl 
 
 ft 
 
 
 n 
 
 
 
 
 3 2 
 
 ;j 
 
 
 -3- 
 H 
 
 CM 
 
 
 vo 
 
 CM 
 
 vo 
 
 CO 
 CO 
 
 
 ^ EH 
 
 CM 
 
 
 
 
 
 
 CM 
 
 CM 
 
 OJ 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 o\ 
 
 CO 
 
 
 
 
 
 
 
 
 
 H 
 
 cu cu 
 
 
 
 
 
 
 
 
 
 
 3 -p 
 
 OO 
 
 CO 
 
 VO 
 
 O 
 
 CO 
 
 t 
 
 LTN 
 
 CM 
 
 fl 
 
 fl d 
 
 ^j- 
 
 vo 
 
 CO 
 
 
 C 
 
 CM 
 
 00 
 
 VO 
 
 W 
 
 cu Q 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 > 1 
 
 VO 
 
 00 
 
 CO 
 
 O 
 
 o 
 
 OO 
 
 O 
 
 OO 
 
 fC 
 
 0) O 
 
 OO 
 
 vo 
 
 OJ 
 
 
 LTN 
 
 H 
 
 CO 
 
 ON 
 
 <c 
 
 EH 
 
 UN 
 
 LTN 
 
 vo 
 
 vo 
 
 OJ 
 
 _-3~ 
 
 vo 
 
 O 
 
 ^>j 
 
 1 
 
 fi 
 
 fl 
 
 fl 
 
 fi 
 
 
 ^ 
 
 fi 
 
 
 
 
 
 ON 
 
 t 
 
 vo 
 
 H 
 
 
 VO 
 
 LT\ 
 
 OJ 
 
 
 03 
 
 
 
 LTN 
 
 H 
 
 
 VO 
 
 CM 
 
 ON 
 
 
 
 
 "* 
 
 
 
 
 
 
 LTv 
 
 Lf\ 
 
 
 
 LfN 
 
 ON 
 
 OO 
 
 
 
 CM 
 
 H 
 
 OO 
 
 
 -p 
 
 00 
 
 00 
 
 OJ 
 
 O 
 
 
 CO 
 
 LTN 
 
 OO 
 
 
 o 
 
 O 
 
 CO 
 
 OO 
 
 vo 
 
 
 O 
 
 OO 
 
 _-.p 
 
 
 a 
 
 
 t 
 
 OO 
 
 OO 
 
 
 CO 
 
 CM 
 
 o 
 
 
 
 J- 
 
 H 
 
 _3~ 
 
 OO 
 
 
 ON 
 
 J- 
 
 j- 
 
 
 CO 
 
 ft 
 
 * 
 
 *s 
 
 ** 
 
 
 n 
 
 fi 
 
 ** 
 
 
 H 
 
 H 
 
 H 
 
 OO 
 
 _3~ 
 
 
 CO 
 
 
 
 ON 
 
 
 rj 
 
 UN 
 
 
 H 
 
 
 
 H 
 
 
 00 
 
 
 ^ 
 
 OJ 
 
 
 
 
 
 
 OJ 
 
 CM 
 
 
 
 
 i 
 
 H 
 
 
 
 
 
 
 
 
 
 
 08 
 
 
 
 
 
 
 
 
 
 
 $ 
 
 
 CO 
 
 
 
 
 
 
 
 ><{ 
 
 EH 
 
 
 3 
 
 
 
 d 
 
 
 
 
 cd 
 
 
 
 o 
 
 
 
 fl 
 
 
 
 
 -p 
 
 -p 
 
 
 cu 
 
 
 
 3 
 
 
 
 
 
 
 
 G 
 
 
 ^ 
 
 pr, 
 
 
 
 
 J>J 
 
 CU 
 
 w 
 
 0) 
 
 -P 
 
 ^i 
 
 
 
 
 
 I_J 
 
 ^ 
 
 CO 
 
 H 
 
 CO 
 
 f-j 
 
 , j 
 
 H^ 
 
 
 
 y^ 
 
 0< 
 
 EH 
 
 H 
 
 cu 
 
 .p 
 
 cd 
 
 <3* 
 
 
 
 0) 
 
 H 
 
 O 
 
 CU 
 
 
 
 IH 
 
 cu 
 
 ^ 
 
 cu 
 
 8 
 
 
 
 o 
 
 H 
 
 -p 
 
 CO 
 
 -p 
 
 
 
 fl 
 
 
 
 
 CM 
 
 ,3 
 
 
 a 1 
 
 fl 
 
 H 
 
 H 
 
 cu 
 O 
 
 
 
 
 00 
 
 O 
 
 CO 
 
 
 
 
 vo 
 
 
 
 
 
 
 cu 
 
 H 
 
 o 
 
 CM 
 
 
 
 -g 
 
 SI 
 
 CM 
 CO 
 
 ON 
 
 
 
 3 
 
 fl 
 
 fi 
 
 . 
 
 
 
 H 
 
 CO 
 
 H 
 
 ON 
 
 
 
 -p 
 
 
 
 CO 
 
 CO 
 
 
 
 CO 
 
 
 00 
 
 t 
 
 
 
 
 OJ 
 
 
 OJ 
 
 
 
 tj 
 
 
 
 
 
 
 cu 
 
 
 
 
 
 
 d 
 
 
 
 
 
 
 fe^. CU 
 
 5? 
 
 t 
 
 OO 
 
 
 
 
 OJ 
 
 H 
 
 OJ 
 
 
 
 
 
 
 
 Q 
 
 
 JH f? 
 cd -g 
 
 CO 
 
 vo 
 
 UN 
 
 H 
 
 OO 
 CO 
 
 J 
 
 
 " i 
 
 H 
 H 
 
 00 
 t 
 
 ON 
 CO 
 
 CU 
 
 <u 
 
 
 -^ co 
 
 -^ 
 
 ON 
 
 fl 
 
 OO 
 
 ~ 
 
 
 a 
 
 OJ 
 
 O 
 
 OO 
 
 u. 
 
 
 
 OO 
 
 CM 
 
 UN 
 
 >2 
 
 
 ^ EH 
 
 OJ 
 
 
 OJ 
 
 
 
 0^ 
 <D 
 
 
 
 
 
 8. 
 
 ^_, 
 
 
 
 
 
 <u 
 
 2 
 
 cu 
 
 
 
 
 06 
 
 -P 
 
 -P 
 
 CM 
 
 _cj- 
 
 VO 
 
 
 H 
 
 o3 
 Q 
 1 
 
 O 
 CM 
 
 t-^- 
 
 UN 
 
 ON 
 
 3 
 C 
 
 CU 
 
 O 
 
 H 
 
 CM 
 
 OO 
 
 > 
 
 
 EH 
 
 O 
 
 _cf 
 
 j- 
 
 u 
 
 X 
 
 1 
 
 n 
 
 n 
 
 n 
 
 p/ 
 
 rq 
 
 co 
 
 OO 
 
 00 
 
 vo 
 
 I 
 
 u 
 
 
 cu 
 
 UN 
 
 
 vo 
 
 "5. 
 
 
 H 
 
 
 
 
 1 
 
 
 
 
 
 
 C/5 
 
 
 a 
 
 OO 
 
 ON 
 
 t 
 
 CO 
 
 
 p 
 
 -^ 
 
 UN 
 
 o 
 
 Sf 
 
 
 o 
 
 00 
 
 cd 
 
 OJ 
 
 ifc 
 
 
 a 
 
 ON 
 
 UN 
 
 UN 
 
 
 
 
 O 
 
 VO 
 
 t- 
 
 
 
 co 
 
 fl 
 
 r* 
 
 ** 
 
 
 
 H 
 
 vo 
 
 H 
 
 C 
 
 
 
 
 CO 
 
 CM 
 
 o 
 
 
 
 EH 
 
 H 
 
 
 OJ 
 
 
STATISTICS REPORT 
 March, 1982 
 
 GOAL: COMMUNICATIONS 
 
 
 
 This 
 
 
 
 $ Diff 
 
 Goal 
 
 Function 
 
 Month 
 
 YTD 
 
 LYTD 
 
 erence 
 
 C 
 
 CALLS 
 
 15112 
 
 42996 
 
 36996 
 
 + 16$ 
 
 C 
 
 ILT 
 
 544 
 
 1251 
 
 601 
 
 +33$ 
 
 C 
 
 LDB 
 
 1771 
 
 5844 
 
 7915 
 
 -26$ 
 
 C 
 
 NLDB 
 
 24 
 
 59 
 
 64 
 
 -08$ 
 
 C 
 
 OUTSIDE 
 
 22 
 
 57 
 
 49 
 
 + 16$ 
 
 C 
 
 P OUT/S 
 
 1215 
 
 2216 
 
 783 
 
 + 183$ 
 
 HANDLED 580 
 LOANED 542 
 BORROWED 142 
 
 Networking 
 
 1703 
 1361 
 456 
 
 1609 
 
 626 
 
 293 
 
 GOAL: INFORMATION 
 
 Services 
 
 I 
 
 PROGRAMS 
 
 111 
 
 229 
 
 I 
 
 P PROGS 
 
 4186 
 
 8801 
 
 I 
 
 QA-D 
 
 7927 
 
 19567 
 
 I 
 
 TOURS 
 
 18 
 
 57 
 
 I 
 
 P TOURS 
 
 406 
 
 1292 
 
 RideFinders 
 
 I 
 
 # CALLS 
 
 391 
 
 1189 
 
 I 
 
 # MATCH 
 
 212 
 
 397 
 
 I 
 
 # ADDED 
 
 111 
 
 391 
 
 1278 
 
 312 
 739 
 
 GOAL: RESOURCE 
 
 Services 
 
 +06$ 
 + 117$ 
 +56$ 
 
 229 
 8801 
 19567 
 57 
 1292 
 
 305 
 11298 
 23172 
 25 
 354 
 
 -33$ 
 -22$ 
 -16$ 
 + 128$ 
 +265$ 
 
 -07$ 
 +27$ 
 -47$ 
 
 R 
 R 
 R 
 R 
 R 
 
 CIRC 
 QA-S 
 REC BOR 
 RSERVES 
 USE 
 
 100655 
 23020 
 2425 
 886 
 60241 
 
 266886 
 57083 
 6870 
 4324 
 153664 
 
 279019 
 63619 
 6203 
 7032 
 161572 
 
 -04$ 
 -10$ 
 + 11$ 
 -39$ 
 -05$ 
 
 Fig. 4. Sample Service Indicator Report for the PPLD 
 
CALLS - telephone calls received 
 
 ILT - intra library transfers, books being transferred from 
 branches or main within the district. 
 
 LDB - library's data base, data base search made of library 
 information. 
 
 NLDB - non-library data base, data base searches of other 
 than the library's. 
 
 OUTSIDE - presentations the staff make outside of the 
 library. 
 
 P OUT/S - number of people at presentations outside of the 
 library. 
 
 PROGRAMS - presentations by the staff presented in the 
 library. 
 
 P PROGS - number of people at presentations presented in 
 the library. 
 
 TOURS - number of library tours given. 
 P TOURS - number of people given tours. 
 
 QA-D - questions answered-directory , questions answered for 
 patrons which does not require any kind of searching in 
 materials (i.e. where are the 900's?) 
 
 CIRC - circulation, number of books checked out 
 
 QA-S - questions answered-search, questions asked by 
 patrons requiring searching for the answer. 
 
 REC BOR - reciprocal borrowing, number of items checked out 
 within the district by patrons who do not belong to the 
 district but belong to the Plains and Peaks Regional 
 Library System. 
 
 RSERVES - number of items placed on reserve. 
 USE - number of people who come in the door. 
 RideFinders: # CALLS - number of telephone calls received. 
 
 # ADDED - number of people inserted in the computer for 
 matching . 
 
 # MATCH - number of people matched to share rides. 
 
 Fig. 4a. Codes for the Service Indicator Report 
 
March, 1982 
 
 IHVKHTORY 
 
 TOTAL MONTHLY 
 
 TOTAL INVENTORY 
 
 THIS YEAR 
 TO DATE 
 
 LAST YEAR 
 TO DATE 
 
 
 + 
 
 _ 
 
 
 
 
 - 
 
 + 
 
 - 
 
 Books (adult & youth) 
 
 2,057 
 
 7U8 
 
 302,258 
 
 5,500 
 
 1,865 
 
 5,021 
 
 2,301 
 
 Books (Juvenile) 
 
 119 
 
 1*21 
 
 98,068 
 
 2,555 
 
 1,633 
 
 1,357 
 
 1,057 
 
 Books (popular) 
 
 226 
 
 85 
 
 Ht, 299 
 
 1*62 
 
 33U 
 
 879 
 
 391* 
 
 Films (8mm) 
 
 
 
 273 
 
 
 
 
 
 Films (16mm) 
 
 
 
 119 
 
 
 1 
 
 
 
 Manuscripts 
 
 
 
 1*31 
 
 
 
 
 
 Map Case, Blueprints 
 
 
 
 2,330 
 
 
 
 
 
 Maps 
 
 
 
 769 
 
 
 
 
 
 Microfiche 
 
 
 
 5,953 
 
 
 
 
 
 Negatives (L.H. ) 
 
 
 
 
 
 
 
 
 Pamphlets, CO 
 
 
 
 265 
 
 
 
 
 
 Prints, Framed 
 
 
 
 332 
 
 
 : 
 
 
 
 Recordings 
 
 
 
 5,192 
 
 
 129 
 
 105 
 
 11*6 
 
 Slides (35mm Color) 
 
 
 
 312 
 
 
 
 
 
 Tapes 
 
 1* kits 
 
 
 139 (125 k) 
 
 16k 
 
 
 
 
 Transparencies 
 
 
 
 21 
 
 
 
 
 
 SUBTOTALS 
 
 2,1*3^ 
 
 1,251* 
 
 1*30,761 
 
 8,566 
 
 3,963 
 
 7,1*85 
 
 3,898 
 
 Microfilm Reels 
 
 
 
 5,517 
 
 
 
 
 
 Vertical File 
 
 776 
 
 38 
 
 31* ,71*2 
 
 1,592 
 
 2,31*6 
 
 2,559 . 
 
 2,898 
 
 Pictures 
 
 166 
 
 108 
 
 37,035 
 
 271* 
 
 127 
 
 273 
 
 91 
 
 Vertical File Music 
 
 
 
 2,01*2 
 
 
 
 
 
 Curiosities 
 
 
 
 51 
 
 
 
 
 
 TOTALS 
 
 3,376 
 
 1,1*00 
 
 510,11*8 
 
 10,1*32 
 
 6,1*36 
 
 1.0,317 
 
 6,887 
 
 OTHER ITEMS OF INTEREST THAT ARE PROCESSED 
 
 Gin Books 
 
 1.06 
 
 
 
 880 
 
 
 69U 
 
 
 Gift Records 
 
 
 
 
 
 
 7 
 
 31 
 
 Bindery Books 
 
 302 
 
 
 
 618 
 
 
 
 
 Priorities 
 
 381 
 
 
 
 1,317 
 
 
 
 
 Government Documents 
 
 1*5 
 
 
 
 52 
 
 
 123 
 
 
 Leased Books 
 
 361 
 
 573 
 
 
 1,1*58 
 
 1,535 
 
 
 
 Periodicals 
 
 910 
 
 
 
 
 
 
 
 Mending Books 
 
 219 
 
 
 
 21*1 
 
 
 
 
 Fig. 5. Sample Inventory Count Report for the PPLD 
 
COMPONENT 
 
 Materials 
 
 Information 
 
 ESTIMATE 
 
 50$ 
 
 3055 
 
 GOAL 
 
 51* 
 
 AGREE 
 
 DISAGREE 
 
 DON'T 
 KNOW 
 
 Communi cat ions 10% 51$ 3 1 ** 5! 
 
 Efficiency 50$ 51* 73* ^ 
 
 Fig. 6. Perceptual Survey Figures for the PPLD 
 
 DATE 
 
 MEAN 
 
 STD EXCPLUS EXCMINUS 
 
 010 
 040 
 060 
 310 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.09^133 
 
 000 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 510 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 560 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 650 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 690 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 770 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 E 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 H 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 L 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 N 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 P 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 Q 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 S 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 X 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 Y 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 Z 
 
 19-APR-82 
 
 0. 
 
 118782 
 
 0.094133 
 
 
 c 
 
 G 
 I 
 
 R 
 
 T 
 
 Fig. 7. Sample Inventory Use Ratio Exceptions Report for the PPLD 
 
DECISION SUPPORT SYSTEM 43 
 
 the number of items in the class. The heading OT (see fig. 8) indicates this 
 number. The OTR heading in figure 8 is the inventory use ratio adjusted 
 for noncirculating items. The results are stored in a datatrieve (the report 
 writer provided by Digital Equipment Corporation) file, from which the 
 report is printed. Records in this file can be retrieved by an element, sorted, 
 formatted for printing, and printed. The results can be arranged by any of 
 the elements. The ability to access this file, as well as the other elements in 
 the DSS via the terminal, allows messaging of the data and the output. 
 Figure 8 shows a report of the inventory use ratio. This program has added 
 a new dimension to our ability to analyze collection usage. It is, perhaps, 
 the best collection planning tool available. It has been used to test the 
 ability of the librarians in information services and reference to estimate 
 the materials needed for the public. Their predictions as to the use of the 
 different classes correlated with the inventory use ratio 80 percent of the 
 time. A ratio for each class can be displayed or printed which allows the 
 trends to be discerned (see fig. 8a). Figure 8b shows the inventory use ratio 
 for new books, and figure 8c shows the inventory use ratio for the 590s. 
 
 Flow Through Expenditures. This report shows the revenues and 
 expenditures for all nongeneral funds. The expenditures in these pro- 
 grams are directly tied to the sources of the funds, and it is important that 
 expenditures do not occur prior to receipt of funds. Examples are grants, 
 federal employment programs, donations and programs that must recover 
 their cost. 
 
 Patron Use. This program checks the date of last use for each patron 
 record and tallies them by class. The classes are less than 30 days, 60 days, 
 1 80 days, one year, two years, three years, four years, and five years. A report 
 is shown in figure 9. This program is run on demand. 
 
 Patron Tenure. This program compares the date the patron received 
 his card with its most recent use. The results indicate the number of current 
 users who received their cards within ninety days, one year, two years, three 
 years, four years, and five years. The results are shown in figure 10. This 
 program is run on demand. 
 
 Security Reports. These reports are daily logs of unusual occurrences 
 in the library. They include the number of patrons trapped by the book 
 security system and other security disturbances. 
 
 Online Catalog Use Count. This is a command log by type and 
 terminal for activity on the online catalog and inventory system. It is 
 printed out daily and accumulated monthly. Since the online catalog is 
 new, this program was established to provide data on its use. A sample of 
 the program is in figure 11. 
 
DATE 
 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19_Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 
 TYPE CLASS 
 
 DEWEY 
 DEWEY 
 SHELF 
 SHELF 
 SHELF 
 SHELF 
 SHELF 
 DEWEY 
 SHELF 
 DEWEY 
 DEWEY 
 SHELF 
 DEWEY 
 SHELF 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 SHELF 
 DEWEY 
 SHELF 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 
 040 
 060 
 T 
 
 G 
 
 C 
 
 R 
 
 310 
 
 I 
 
 010 
 
 050 
 
 B 
 
 860 
 
 K 
 
 840 
 
 960 
 
 320 
 
 090 
 
 850 
 
 920 
 
 080 
 
 500 
 
 400 
 
 030 
 
 880 
 
 870 
 
 380 
 
 820 
 
 830 
 
 890 
 
 550 
 
 570 
 
 980 
 
 340 
 
 V 
 
 900 
 
 A 
 
 020 
 
 260 
 
 270 
 
 970 
 
 350 
 
 190 
 
 660 
 
 540 
 
 700 
 
 710 
 
 990 
 
 810 
 
 TTL 
 
 4 
 
 92 
 
 16 
 
 
 
 2,952 
 11,162 
 13,827 
 1,141 
 173 
 1,372 
 238 
 3,505 
 598 
 63 
 785 
 297 
 3,678 
 53 
 226 
 14,211 
 85 
 ,610 
 101 
 ,427 
 239 
 107 
 3,131 
 4,364 
 769 
 731 
 3,018 
 2,320 
 
 1 
 
 1, 
 
 1 
 
 1 
 
 253 
 ,356 
 50 
 ,148 
 16 
 
 ,709 
 522 
 330 
 
 13,368 
 
 4,400 
 
 446 
 
 515 
 
 829 
 
 1,831 
 
 655 
 
 128 
 
 7,921 
 
 20-Apr-82 
 
 Page 1 
 
 OUT 
 
 REF 
 
 OT 
 
 OTR 
 
 SUM 
 
 
 
 1 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 
 
 44 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 
 
 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 
 
 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 5 
 
 
 
 0.0017 
 
 0.0000 
 
 0.0000 
 
 20 
 
 
 
 0.0018 
 
 0.0000 
 
 0.0000 
 
 H2 
 
 
 
 0.0030 
 
 0.0000 
 
 0.0000 
 
 9 
 
 998 
 
 0.0079 
 
 0.0629 
 
 0.0000 
 
 2 
 
 
 
 0.0116 
 
 0.0000 
 
 0.0000 
 
 16 
 
 1,042 
 
 0.0117 
 
 0.0485 
 
 0.0000 
 
 6 
 
 192 
 
 0.0252 
 
 0.1304 
 
 0.0000 
 
 89 
 
 
 
 0.0254 
 
 0.0000 
 
 0.0000 
 
 18 
 
 5 
 
 0.0301 
 
 0.0304 
 
 0.0000 
 
 2 
 
 
 
 0.0317 
 
 0.0000 
 
 0.0000 
 
 25 
 
 13 
 
 0.0318 
 
 0.0324 
 
 0.0000 
 
 10 
 
 3 
 
 0.0337 
 
 0.0340 
 
 0.0000 
 
 125 
 
 438 
 
 0.0340 
 
 0.0386 
 
 0.0000 
 
 2 
 
 7 
 
 0.0377 
 
 0.0435 
 
 0.0000 
 
 9 
 
 3 
 
 0.0398 
 
 0.0404 
 
 0.0000 
 
 617 
 
 2,729 
 
 0.0434 
 
 0.0537 
 
 0.0000 
 
 4 
 
 4 
 
 0.0471 
 
 0.0494 
 
 0.0000 
 
 78 
 
 520 
 
 0.0484 
 
 0.0716 
 
 0.0000 
 
 5 
 
 1 
 
 0.0495 
 
 0.0500 
 
 0.0000 
 
 75 
 
 925 
 
 0.0526 
 
 0.1494 
 
 0.0000 
 
 13 
 
 4 
 
 0.0544 
 
 0.0553 
 
 0.0000 
 
 6 
 
 2 
 
 0.0561 
 
 0.0571 
 
 0.0000 
 
 177 
 
 1,107 
 
 0.0565 
 
 0.0875 
 
 0.0000 
 
 21*9 
 
 314 
 
 0.0571 
 
 0.0615 
 
 0.0000 
 
 44 
 
 24 
 
 0.0572 
 
 0.0591 
 
 0.0000 
 
 42 
 
 19 
 
 0.0575 
 
 0.0590 
 
 0.0000 
 
 174 
 
 1,299 
 
 0.0577 
 
 0.1012 
 
 0.0000 
 
 136 
 
 356 
 
 0.0586 
 
 0.0692 
 
 0.0000 
 
 15 
 
 3 
 
 0.0593 
 
 0.0600 
 
 0.0000 
 
 200 
 
 1,087 
 
 0.0596 
 
 0.0881 
 
 0.0000 
 
 3 
 
 
 
 0.0600 
 
 0.0000 
 
 0.0000 
 
 69 
 
 162 
 
 0.0601 
 
 0.0700 
 
 0.0000 
 
 1 
 
 
 
 0.0625 
 
 0.0000 
 
 0.0000 
 
 108 
 
 556 
 
 0.0632 
 
 0.0937 
 
 0.0000 
 
 33 
 
 61 
 
 0.0632 
 
 0.0716 
 
 0.0000 
 
 21 
 
 43 
 
 0.0636 
 
 0.0732 
 
 0.0000 
 
 857 
 
 2,635 
 
 0.0641 
 
 0.0798 
 
 0.0000 
 
 289 
 
 1,176 
 
 0.0657 
 
 0.0896 
 
 0.0000 
 
 30 
 
 6 
 
 0.0673 
 
 0.0682 
 
 0.0000 
 
 35 
 
 56 
 
 0.0680 
 
 0.0763 
 
 0.0000 
 
 57 
 
 175 
 
 0.0688 
 
 0.0872 
 
 0.0000 
 
 128 
 
 192 
 
 0.0699 
 
 0.0781 
 
 0.0000 
 
 46 
 
 266 
 
 0.0702 
 
 0.1183 
 
 0.0000 
 
 9 
 
 
 
 0.0703 
 
 0.0703 
 
 0.0000 
 
 560 
 
 560 
 
 0.0707 
 
 0.0761 
 
 0.0000 
 
 Fig. 8. Sample Inventory Use Ratio Report for the PPLD 
 
DATE 
 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19_Apr- 82 
 19-Apr-82 
 19_Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19-Apr-82 
 19_Apr-82 
 19-Apr-82 
 
 TYPE CLASS 
 
 SHELF 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 DEWEY 
 
 000 
 010 
 020 
 030 
 040 
 050 
 060 
 070 
 080 
 090 
 100 
 110 
 120 
 130 
 140 
 150 
 160 
 170 
 180 
 190 
 200 
 210 
 220 
 230 
 240 
 250 
 260 
 270 
 280 
 290 
 300 
 310 
 320 
 330 
 340 
 350 
 360 
 370 
 380 
 390 
 400 
 410 
 420 
 430 
 440 
 450 
 460 
 470 
 
 TTL 
 
 
 
 1,004 
 1,372 
 1,709 
 1,427 
 4 
 
 238 
 
 92 
 
 613 
 
 85 
 
 53 
 
 157 
 
 127 
 
 253 
 
 2,296 
 
 116 
 
 2,401 
 
 58 
 
 518 
 
 330 
 
 446 
 
 381 
 
 133 
 
 1,147 
 
 787 
 
 991 
 
 209 
 
 522 
 
 330 
 
 646 
 
 1,527 
 
 5,474 
 
 1,141 
 
 3,678 
 
 5,691 
 
 3,356 
 
 4,400 
 
 5,252 
 
 4,122 
 
 3,131 
 
 7,327 
 
 101 
 
 207 
 
 874 
 
 123 
 
 128 
 
 27 
 
 262 
 
 21 
 
 20-Apr-82 
 
 Page 1 
 
 OUT 
 
 REF 
 
 OT 
 
 OTR 
 
 SUM 
 
 
 
 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 233 
 
 38 
 
 0.2321 
 
 0.2412 
 
 0.0000 
 
 16 
 
 1,042 
 
 0.0117 
 
 0.0485 
 
 0.0000 
 
 108 
 
 556 
 
 0.0632 
 
 0.0937 
 
 0.0000 
 
 75 
 
 925 
 
 0.0526 
 
 0.1494 
 
 0.0000 
 
 
 
 1 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 6 
 
 192 
 
 0.0252 
 
 0.1304 
 
 0.0000 
 
 
 
 44 
 
 0.0000 
 
 0.0000 
 
 0.0000 
 
 44 
 
 173 
 
 0.0718 
 
 0.1000 
 
 0.0000 
 
 4 
 
 4 
 
 0.0471 
 
 0.0494 
 
 0.0000 
 
 2 
 
 7 
 
 0.0377 
 
 0.0435 
 
 0.0000 
 
 23 
 
 17 
 
 0.1465 
 
 0.1643 
 
 0.0000 
 
 13 
 
 6 
 
 0.1024 
 
 0.1074 
 
 0.0000 
 
 23 
 
 1 
 
 0.0909 
 
 0.0913 
 
 0.0000 
 
 471 
 
 23 
 
 0.2051 
 
 0.2072 
 
 0.0000 
 
 9 
 
 
 
 0.0776 
 
 0.0776 
 
 0.0000 
 
 490 
 
 14 
 
 0.2041 
 
 0.2053 
 
 0.0000 
 
 9 
 
 1 
 
 0.1552 
 
 0.1579 
 
 0.0000 
 
 64 
 
 23 
 
 0.1236 
 
 0.1293 
 
 0.0000 
 
 42 
 
 1 
 
 0.1273 
 
 0.1277 
 
 0.0000 
 
 30 
 
 6 
 
 0.0673 
 
 0.0682 
 
 0.0000 
 
 41 
 
 59 
 
 0.1076 
 
 0.1273 
 
 0.0000 
 
 14 
 
 7 
 
 0.1053 
 
 0.1111 
 
 0.0000 
 
 128 
 
 69 
 
 0.1116 
 
 0.1187 
 
 0.0000 
 
 75 
 
 64 
 
 0.0953 
 
 0.1037 
 
 0.0000 
 
 145 
 
 41 
 
 0.1463 
 
 0.1526 
 
 0.0000 
 
 17 
 
 19 
 
 0.0813 
 
 0.0895 
 
 0.0000 
 
 33 
 
 61 
 
 0.0632 
 
 0.0716 
 
 0.0000 
 
 21 
 
 43 
 
 0.0636 
 
 0.0732 
 
 0.0000 
 
 66 
 
 126 
 
 0.1022 
 
 0.1269 
 
 0.0000 
 
 178 
 
 94 
 
 0.1166 
 
 0.1242 
 
 0.0000 
 
 504 
 
 525 
 
 0.0921 
 
 0.1018 
 
 0.0000 
 
 9 
 
 998 
 
 0.0079 
 
 0.0629 
 
 0.0000 
 
 125 
 
 438 
 
 0.0340 
 
 0.0386 
 
 0.0000 
 
 551 
 
 1,216 
 
 0.0968 
 
 0.1231 
 
 0.0000 
 
 200 
 
 1,087 
 
 0.0596 
 
 0.0881 
 
 0.0000 
 
 289 
 
 1,176 
 
 0.0657 
 
 0.0896 
 
 0.0000 
 
 507 
 
 1,171 
 
 0.0965 
 
 0.1242 
 
 0.0000 
 
 446 
 
 838 
 
 0.1082 
 
 0.1358 
 
 0.0000 
 
 177 
 
 1,107 
 
 0.0565 
 
 0.0875 
 
 0.0000 
 
 524 
 
 353 
 
 0.0715 
 
 0.0751 
 
 0.0000 
 
 5 
 
 1 
 
 0.0495 
 
 0.0500 
 
 0.0000 
 
 33 
 
 6 
 
 0.1594 
 
 0.1642 
 
 0.0000 
 
 104 
 
 163 
 
 0.1190 
 
 0.1463 
 
 0.0000 
 
 19 
 
 18 
 
 0.1545 
 
 0.1810 
 
 0.0000 
 
 21 
 
 7 
 
 0.1641 
 
 0.1736 
 
 0.0000 
 
 5 
 
 3 
 
 0.1852 
 
 0.2083 
 
 0.0000 
 
 43 
 
 17 
 
 0.1641 
 
 0.1755 
 
 0.0000 
 
 3 
 
 3 
 
 0.1429 
 
 0.1667 
 
 0.0000 
 
 Fig. 8a. Sample Inventory Use Ratio by Class Report for the PPLD 
 
DATE 
 
 TYPE CLASS 
 
 TTL 
 
 OUT 
 
 REF 
 
 OT 
 
 OTR 
 
 SUM 
 
 29-Feb-80 SHELF 
 
 N 
 
 2,008 
 
 1,262 
 
 04-Apr-80 SHELF 
 
 N 
 
 2,455 
 
 1,435 
 
 05-May-80 SHELF 
 
 N 
 
 2,097 
 
 1,253 
 
 31_May-80 SHELF 
 
 N 
 
 3,072 
 
 1,833 
 
 01-Jul-80 SHELF 
 
 N 
 
 3,^73 
 
 2,153 
 
 01-Aug-80 SHELF 
 
 N 
 
 3,554 
 
 2,122 
 
 03-Sep-80 SHELF 
 
 N 
 
 3,917 
 
 2,172 
 
 05-0ct-80 SHELF 
 
 N 
 
 4,608 
 
 2,764 
 
 03-Nov-80 SHELF 
 
 N 
 
 5,384 
 
 3,220 
 
 30-Nov-80 SHELF 
 
 N 
 
 6,426 
 
 3,635 
 
 01-Jan-8l SHELF 
 
 N 
 
 7,515 
 
 4,102 
 
 01-Feb-8l SHELF 
 
 N 
 
 7,920 
 
 3,662 
 
 01-Mar-8l SHELF 
 
 N 
 
 8,836 
 
 4,637 
 
 05-Apr-8l SHELF 
 
 N 
 
 9,488 
 
 5,099 
 
 11-May-8l SHELF 
 
 N 
 
 9,488 
 
 5,099 
 
 01-Jun-8l SHELF 
 
 N 
 
 10,007 
 
 5,185 
 
 09-JU1-81 SHELF 
 
 N 
 
 9,969 
 
 5,152 
 
 31-Aug-8l SHELF 
 
 N 
 
 9,877 
 
 4,766 
 
 15-Feb-82 SHELF 
 
 N 
 
 10,945 
 
 5,619 
 
 02-Mar-82 SHELF 
 
 N 
 
 10,944 
 
 5,740 
 
 19_Apr-82 SHELF 
 
 N 
 
 9,877 
 
 4,766 
 
 
 
 0.6285 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5845 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5975 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5967 
 
 0.0000 
 
 0.0000 
 
 
 
 0.6199 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5971 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5545 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5998 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5981 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5657 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5458 
 
 0.0000 
 
 0.0000 
 
 
 
 0.4624 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5248 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5374 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5374 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5181 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5168 
 
 0.0000 
 
 0.0000 
 
 
 
 0.4825 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5134 
 
 0.0000 
 
 0.0000 
 
 
 
 0.5245 
 
 0.0000 
 
 0.0000 
 
 
 
 0.4825 
 
 0.0000 
 
 0.0000 
 
 Fig. 8b. Sample Inventory Use Ratio Report for New Books for the PPLD 
 
 DATE 
 
 TYPE CLASS TTL 
 
 OUT 
 
 REF 
 
 OT 
 
 OTR 
 
 SUM 
 
 29-Feb-80 
 
 DEWEY 
 
 590 
 
 4 
 
 ,379 
 
 658 
 
 29 
 
 0. 
 
 1503 
 
 0. 
 
 1513 
 
 0.0000 
 
 04-Apr-80 
 
 DEWEY 
 
 590 
 
 4 
 
 ,538 
 
 817 
 
 30 
 
 0. 
 
 1800 
 
 0. 
 
 1812 
 
 0.0000 
 
 05-May-80 
 
 DEWEY 
 
 590 
 
 4 
 
 ,388 
 
 668 
 
 29 
 
 0. 
 
 1522 
 
 0. 
 
 1532 
 
 0.0000 
 
 31-May-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,115 
 
 636 
 
 152 
 
 0. 
 
 1243 
 
 0. 
 
 1281 
 
 0.0000 
 
 01-Jul-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,184 
 
 603 
 
 172 
 
 0. 
 
 1163 
 
 0. 
 
 1203 
 
 0.0000 
 
 01-Aug-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,249 
 
 653 
 
 172 
 
 0. 
 
 1244 
 
 0. 
 
 1286 
 
 0.0000 
 
 03-Sep-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,298 
 
 565 
 
 173 
 
 0. 
 
 1066 
 
 0. 
 
 1102 
 
 0.0000 
 
 05-0ct-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,394 
 
 882 
 
 174 
 
 0. 
 
 1635 
 
 0. 
 
 1690 
 
 0.0000 
 
 03-Nov-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,498 
 
 833 
 
 174 
 
 0. 
 
 1515 
 
 0. 
 
 1565 
 
 0.0000 
 
 30-Nov-80 
 
 DEWEY 
 
 590 
 
 5 
 
 ,534 
 
 794 
 
 174 
 
 0. 
 
 1435 
 
 0. 
 
 1481 
 
 0.0000 
 
 01-Jan-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,632 
 
 552 
 
 174 
 
 0. 
 
 0980 
 
 0. 
 
 1011 
 
 0.0000 
 
 01-Feb-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,685 
 
 713 
 
 200 
 
 0. 
 
 1254 
 
 0. 
 
 1300 
 
 0.0000 
 
 01-Mar-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,716 
 
 1 ,008 
 
 201 
 
 0. 
 
 1763 
 
 0. 
 
 1828 
 
 0.0000 
 
 05-Apr-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,873 
 
 1,134 
 
 250 
 
 0. 
 
 1931 
 
 0. 
 
 2017 
 
 0.0000 
 
 11-May-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,873 
 
 1 ,134 
 
 250 
 
 0. 
 
 1931 
 
 0. 
 
 2017 
 
 0.0000 
 
 01-Jun-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,939 
 
 820 
 
 255 
 
 0. 
 
 1381 
 
 0. 
 
 1443 
 
 0.0000 
 
 09-JU1-81 
 
 DEWEY 
 
 590 
 
 5 
 
 ,976 
 
 829 
 
 253 
 
 0. 
 
 1387 
 
 0. 
 
 1449 
 
 0.0000 
 
 31-Aug-8l 
 
 DEWEY 
 
 590 
 
 5 
 
 ,898 
 
 674 
 
 253 
 
 0. 
 
 1143 
 
 0. 
 
 1194 
 
 0.0000 
 
 15-Feb-82 
 
 DEWEY 
 
 590 
 
 6 
 
 ,093 
 
 924 
 
 256 
 
 0. 
 
 1516 
 
 0. 
 
 1583 
 
 0.0000 
 
 02-Mar-82 
 
 DEWEY 
 
 590 
 
 6 
 
 ,100 
 
 926 
 
 256 
 
 0. 
 
 1518 
 
 0. 
 
 1585 
 
 0.0000 
 
 19-Apr-82 
 
 DEWEY 
 
 590 
 
 5 
 
 ,898 
 
 674 
 
 253 
 
 0. 
 
 1143 
 
 0. 
 
 1194 
 
 0.0000 
 
 Fig. 8c. Sample Inventory Use Ratio Report for the 590 Class for the PPLD 
 
UDATE 
 
 USEX 
 
 UDESIG 
 
 UCLASS 
 
 UTOTAL 
 
 UUSE 
 
 lU-Mar-82 
 
 M 
 
 B 
 
 30 
 
 000069 
 
 000.0008 
 
 lU-Mar-82 
 
 M 
 
 M 
 
 30 
 
 000023 
 
 000.0003 
 
 lU-Mar-82 
 
 M 
 
 P 
 
 30 
 
 002652 
 
 000.0291 
 
 lU-Mar-82 
 
 W 
 
 B 
 
 30 
 
 00020U 
 
 000.0022 
 
 lU-Mar-82 
 
 W 
 
 M 
 
 30 
 
 000077 
 
 000.0008 
 
 lU-Mar-82 
 
 W 
 
 P 
 
 30 
 
 003782 
 
 OOO.OU15 
 
 lU-Mar-82 
 
 B 
 
 B 
 
 30 
 
 000027 
 
 000.0003 
 
 lU-Mar-82 
 
 B 
 
 M 
 
 30 
 
 000019 
 
 000.0002 
 
 lU-Mar-82 
 
 B 
 
 P 
 
 30 
 
 000903 
 
 000.0099 
 
 lU-Mar-82 
 
 G 
 
 B 
 
 30 
 
 OOOOU2 
 
 000.0005 
 
 lU-Mar-82 
 
 G 
 
 M 
 
 30 
 
 000017 
 
 000.0002 
 
 lU-Mar-82 
 
 G 
 
 P 
 
 30 
 
 001079 
 
 000.0118 
 
 lU-Mar-82 
 
 U 
 
 B 
 
 30 
 
 000009 
 
 000.0001 
 
 lU-Mar-82 
 
 U 
 
 M 
 
 30 
 
 000005 
 
 000.0001 
 
 lU-Mar-82 
 
 U 
 
 P 
 
 30 
 
 000768 
 
 000.008U 
 
 lU-Mar-82 
 
 M 
 
 B 
 
 90 
 
 000051 
 
 000.0006 
 
 lU-Mar-82 
 
 M 
 
 M 
 
 90 
 
 000020 
 
 000.0002 
 
 lU-Mar-82 
 
 M 
 
 P 
 
 90 
 
 002623 
 
 000.0288 
 
 lH-Mar-82 
 
 W 
 
 B 
 
 90 
 
 000096 
 
 000.0011 
 
 lU-Mar-82 
 
 W 
 
 M 
 
 90 
 
 OOOOUO 
 
 OOO.OOOU 
 
 lU-Mar-82 
 
 W 
 
 P 
 
 90 
 
 003237 
 
 000.0355 
 
 lU-Mar-82 
 
 B 
 
 B 
 
 90 
 
 00002U 
 
 000.0003 
 
 lU-Mar-82 
 
 B 
 
 M 
 
 90 
 
 000019 
 
 000.0002 
 
 lU-Mar-82 
 
 B 
 
 P 
 
 90 
 
 000881 
 
 000.0097 
 
 lU-Mar-82 
 
 G 
 
 B 
 
 90 
 
 000039 
 
 OOO.OOOU 
 
 lU-Mar-82 
 
 G 
 
 M 
 
 90 
 
 00002U 
 
 000.0003 
 
 lU-Mar-82 
 
 G 
 
 P 
 
 90 
 
 00099^ 
 
 000.0109 
 
 lU-Mar-82 
 
 U 
 
 B 
 
 90 
 
 000008 
 
 000.0001 
 
 lU-Mar-82 
 
 U 
 
 M 
 
 90 
 
 000002 
 
 000.0000 
 
 lU-Mar-82 
 
 U 
 
 P 
 
 90 
 
 000588 
 
 000.006U 
 
 lU-Mar-82 
 
 M 
 
 B 
 
 180 
 
 ooooUo 
 
 OOO.OOOU 
 
 lU-Mar-82 
 
 M 
 
 M 
 
 180 
 
 000012 
 
 000.0001 
 
 lU-Mar-82 
 
 M 
 
 P 
 
 180 
 
 0028UO 
 
 000.0311 
 
 lU-Mar-82 
 
 W 
 
 B 
 
 180 
 
 000125 
 
 OOO.OOlU 
 
 lU-Mar-82 
 
 W 
 
 M 
 
 180 
 
 OOOOUU 
 
 000.0005 
 
 lU-Mar-82 
 
 W 
 
 P 
 
 180 
 
 003831 
 
 OOO.OU20 
 
 lU-Mar-82 
 
 B 
 
 B 
 
 180 
 
 000051 
 
 000.0006 
 
 lU-Mar-82 
 
 B 
 
 M 
 
 180 
 
 000025 
 
 000.0003 
 
 lU-Mar-82 
 
 B 
 
 P 
 
 180 
 
 000992 
 
 000.0109 
 
 lU-Mar-82 
 
 G 
 
 B 
 
 180 
 
 OOOOU8 
 
 000.0005 
 
 lU_Mar-82 
 
 G 
 
 M 
 
 180 
 
 000030 
 
 000.0003 
 
 lU-Mar-82 
 
 G 
 
 P 
 
 180 
 
 0010U7 
 
 000.0115 
 
 lU-Mar-82 
 
 U 
 
 B 
 
 180 
 
 000005 
 
 000.0001 
 
 lU_Mar-82 
 
 U 
 
 M 
 
 180 
 
 000001 
 
 000.0000 
 
 lU-Mar-82 
 
 U 
 
 P 
 
 180 
 
 0005^6 
 
 000.0060 
 
 Fig. 9. Sample Patron Use Report for the PPLD 
 
7000- 
 
 2500 - 
 
 2000- 
 
 1500 - 
 
 1000 
 
 500 
 
 1982 
 
 1981 
 
 I960 
 
 1979 
 
 1978 
 
 1977 
 
 1976 
 
 Year patron received library card 
 
 Fig. 10. Graphed Results of the Patron Tenure Report for the PPLD 
 
KB5: 
 
 Cmdr Co die 
 
 118 
 
 17 
 
 15 
 
 63 
 
 255 
 
 18 
 
 66 
 
 175 
 
 727 
 
 0.76? 
 
 KB6: 
 
 Nirvana 
 
 6 
 
 
 
 
 
 
 
 25 
 
 
 
 
 
 93 
 
 121* 
 
 0.13$ 
 
 KB7: 
 
 Joyne Hunt 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 
 8 
 
 o.oi* 
 
 KB8: 
 
 Rick Richmond 
 
 1 
 
 
 
 
 
 
 
 1* 
 
 
 
 1 
 
 
 
 6 
 
 0.01% 
 
 KB11: 
 
 Sally Palmer 
 
 27 
 
 23 
 
 
 
 10 
 
 200 
 
 6 
 
 6 
 
 206 
 
 1*78 
 
 0.50* 
 
 KB12: 
 
 Emmitt Kraft 
 
 
 
 1 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 2 
 
 0.00? 
 
 KB13: 
 
 AuxSvcs Orders 
 
 1 
 
 5 
 
 
 
 
 
 92 
 
 
 
 
 
 79 
 
 177 
 
 0.19* 
 
 KB16: 
 
 PIO #2 
 
 
 
 
 
 
 
 
 
 1* 
 
 
 
 2 
 
 1 
 
 7 
 
 0.01* 
 
 KB18: 
 
 East Br check 
 
 
 
 193 
 
 1 
 
 
 
 11 
 
 
 
 
 
 11 
 
 216 
 
 0.23* 
 
 KB20: 
 
 Catalog I 
 
 
 
 
 
 1 
 
 
 
 1 
 
 
 
 
 
 
 
 2 
 
 0.00* 
 
 KB21: 
 
 East Br IHFO 
 
 101* 
 
 1*9 
 
 7 
 
 1 
 
 2086 
 
 
 
 297 
 
 778 
 
 3322 
 
 3.1*8* 
 
 KB23: 
 
 AuxSvcs Ibuild 
 
 1 
 
 11* 
 
 
 
 
 
 7 
 
 
 
 2 
 
 8 
 
 32 
 
 0.03* 
 
 KB21*: 
 
 Billing Divn 
 
 
 
 512 
 
 
 
 2 
 
 18 
 
 
 
 1 
 
 7 
 
 51*0 
 
 0.57* 
 
 KB25: 
 
 Core I 
 
 51* 
 
 1 
 
 
 
 2 
 
 831 
 
 
 
 15 
 
 509 
 
 11*12 
 
 1.1.8* 
 
 KB26: 
 
 Patron Regis 
 
 2 
 
 135 
 
 
 
 
 
 53 
 
 
 
 
 
 5 
 
 195 
 
 0.20* 
 
 KB27: 
 
 Book Return 
 
 1 
 
 15 
 
 
 
 
 
 36 
 
 
 
 
 
 2 
 
 51* 
 
 0.06* 
 
 KB28: 
 
 Chargeout E 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 0.00* 
 
 KB29: 
 
 Chargeout W 
 
 3 
 
 
 
 
 
 
 
 7 
 
 
 
 1* 
 
 13 
 
 27 
 
 0.03* 
 
 KB30: 
 
 300 baud dialup 
 
 51 
 
 1* 
 
 8 
 
 56 
 
 556 
 
 2 
 
 60 
 
 32 
 
 769 
 
 0.81* 
 
 KB31: 
 
 300 baud dialup 
 
 33 
 
 1 
 
 2 
 
 16 
 
 382 
 
 
 
 1*5 
 
 58 
 
 537 
 
 0.56* 
 
 KB32: 
 
 300 baud dialup 
 
 16 
 
 1 
 
 
 
 17 
 
 153 
 
 
 
 21* 
 
 9 
 
 220 
 
 0.23* 
 
 KB33: 
 
 300 baud dialup 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 5 
 
 
 
 6 
 
 0.01* 
 
 KB31*: 
 
 300 baud dialup 
 
 i 
 
 
 
 
 
 
 
 3 
 
 
 
 8 
 
 3 
 
 15 
 
 0.02* 
 
 KB 38: 
 
 INFO East 
 
 177 
 
 1 
 
 9 
 
 3 
 
 2U02 
 
 1 
 
 11*8 
 
 699 
 
 31*1*0 
 
 3.60* 
 
 KB39: 
 
 REFER South 
 
 1*9 
 
 2 
 
 8 
 
 11 
 
 1166 
 
 
 
 126 
 
 1*03 
 
 1765 
 
 1.85* 
 
 KBUO: 
 
 N MOD office 
 
 2 
 
 
 
 
 
 1 
 
 191 
 
 
 
 11 
 
 1*80 
 
 685 
 
 0.72* 
 
 KB Ul: 
 
 INFO West 
 
 57 
 
 
 
 ll* 
 
 8 
 
 1335 
 
 
 
 168 
 
 61*9 
 
 2231 
 
 2.31** 
 
 KB 1*3: 
 
 AuxSvcs Mod 1 
 
 
 
 
 
 1 
 
 
 
 71 
 
 
 
 
 
 6 
 
 78 
 
 0.08* 
 
 KBV*: 
 
 Reserves 
 
 11* 
 
 ll*9 
 
 18 
 
 
 
 1128 
 
 
 
 1 
 
 328 
 
 1638 
 
 1-71* 
 
 KB 1*5: 
 
 AuxSvcs Ibuild 
 
 1 
 
 
 
 2 
 
 
 
 22 
 
 
 
 
 
 1* 
 
 29 
 
 0.03* 
 
 KB 1*6: 
 
 Periodicals Off 
 
 2 
 
 1 
 
 
 
 
 
 1*9 
 
 
 
 
 
 56 
 
 108 
 
 0.11* 
 
 KBl*7: 
 
 Reference 
 
 56 
 
 3 
 
 13 
 
 8 
 
 11*81* 
 
 
 
 162 
 
 635 
 
 2361 
 
 2.1*7* 
 
 KB U8: 
 
 Reserves 
 
 25 
 
 73 
 
 8 
 
 
 
 lit 18 
 
 
 
 
 
 1*31* 
 
 1958 
 
 2.05* 
 
 KB 1*9: 
 
 AuxSvcs Ibuild 
 
 
 
 11. 
 
 
 
 
 
 2 
 
 
 
 
 
 2 
 
 18 
 
 0.02* 
 
 KB50: 
 
 Local History 
 
 11 
 
 
 
 
 
 
 
 69 
 
 
 
 12 
 
 25 
 
 117 
 
 0.12* 
 
 KB53: 
 
 Ute Pass Branch 
 
 8 
 
 1 
 
 1 
 
 
 
 95 
 
 
 
 2 
 
 39 
 
 11*6 
 
 0.15* 
 
 KB56: 
 
 Calhan Branch 
 
 1 
 
 21* 
 
 
 
 1 
 
 60 
 
 
 
 2 
 
 35 
 
 123 
 
 0.13* 
 
 KB57: 
 
 Monument Branch 
 
 21 
 
 17 
 
 
 
 
 
 257 
 
 
 
 6 
 
 1*3 
 
 31*1* 
 
 0.36* 
 
 KB58: 
 
 Fountain Branch 
 
 3 
 
 1 
 
 
 
 
 
 111 
 
 
 
 1* 
 
 20 
 
 139 
 
 0.15* 
 
 KB59: 
 
 Old Colo City B 
 
 59 
 
 125 
 
 13 
 
 
 
 1371 
 
 
 
 32 
 
 261* 
 
 1861* 
 
 1.95* 
 
 KB60: 
 
 Broadmkt Branch 
 
 21 
 
 ll* 
 
 
 
 
 
 379 
 
 
 
 5 
 
 129 
 
 51*8 
 
 0.57* 
 
 KB6l: 
 
 East Br CHARGE 
 
 2 
 
 38 
 
 
 
 
 
 27 
 
 
 
 
 
 3 
 
 70 
 
 0.07* 
 
 KB62: 
 
 Palmer Lk Branch 
 
 8 
 
 8 
 
 
 
 
 
 227 
 
 
 
 10 
 
 19 
 
 272 
 
 0.28* 
 
 KB63: 
 
 AuxSvcs Ibuild 
 
 1* 
 
 3 
 
 
 
 
 
 21* 
 
 
 
 
 
 7 
 
 38 
 
 o.oi** 
 
 KB61*: 
 
 Basement Checkln 
 
 2 
 
 1 
 
 
 
 2 
 
 60 
 
 
 
 7 
 
 6 
 
 78 
 
 0.08* 
 
 KB66: 
 
 RideFinders 
 
 1 
 
 
 
 
 
 
 
 1*8 
 
 
 
 3 
 
 ll* 
 
 66 
 
 0.07* 
 
 KB68: 
 
 Mr. Dowlin 
 
 2 
 
 
 
 
 
 10 
 
 26 
 
 
 
 2 
 
 3 
 
 U3 
 
 0.05* 
 
 KB69: 
 
 Gail Stowe 
 
 
 
 
 
 
 
 
 
 109 
 
 
 
 13 
 
 17 
 
 139 
 
 0.15* 
 
 KB72: 
 
 Interlib loan 
 
 5 
 
 5 
 
 1 
 
 3 
 
 21* 
 
 
 
 7 
 
 7 
 
 52 
 
 0.05* 
 
 KB73: 
 
 Outreach 
 
 29 
 
 3 
 
 2 
 
 1 
 
 9l0 
 
 
 
 1*5 
 
 225 
 
 121*5 
 
 1.30* 
 
 KB77: 
 
 Mrs. Sheff 
 
 9 
 
 9 
 
 3 
 
 8 
 
 612 
 
 
 
 11 
 
 3U3 
 
 995 
 
 l.OU* 
 
 Fig. 11. Report of Online Catalog Use Statistics for the PPLD 
 
KEN DOW LIN b LYNN MAGRATH 
 
 Fig. 11 Continued 
 
 KB78: 
 
 Catalog II 
 
 8 
 
 5 
 
 1 
 
 
 
 72 
 
 
 
 
 
 3U 
 
 120 
 
 0.13? 
 
 KB80: 
 
 East Public 1 
 
 235 
 
 7 
 
 U7 
 
 91* 
 
 2738 
 
 10 
 
 91*8 
 
 638 
 
 1*717 
 
 l*.9l*? 
 
 KB81: 
 
 East Public YA 
 
 231 
 
 13 
 
 35 
 
 79 
 
 2261* 
 
 6 
 
 725 
 
 391* 
 
 371*7 
 
 3.92? 
 
 KB82: 
 
 Public Children 
 
 202 
 
 114 
 
 39 
 
 110 
 
 211*5 
 
 10 
 
 526 
 
 581 
 
 3627 
 
 3.80? 
 
 KB83: 
 
 Sys Off spare 
 
 3 
 
 
 
 
 
 
 
 1*6 
 
 
 
 9 
 
 1 
 
 59 
 
 0.06? 
 
 KB85: 
 
 Publi SI 
 
 390 
 
 1*14 
 
 123 
 
 219 
 
 1*71*8 
 
 21* 
 
 1088 
 
 656 
 
 7292 
 
 7-63? 
 
 KB86: 
 
 Publi COCIS 
 
 18 
 
 U 
 
 6 
 
 12 
 
 506 
 
 1 
 
 lUO 
 
 67 
 
 751* 
 
 0.79? 
 
 KB87: 
 
 Publi S3 
 
 1*01* 
 
 21 
 
 150 
 
 11*1 
 
 3698 
 
 13 
 
 101*9 
 
 565 
 
 60 1*1 
 
 6.32? 
 
 KB89: 
 
 Publi N2 
 
 391 
 
 11 
 
 71 
 
 151 
 
 1*982 
 
 19 
 
 1353 
 
 706 
 
 7719 
 
 8.08? 
 
 KB90: 
 
 Publi N3 
 
 293 
 
 27 
 
 61 
 
 108 
 
 3760 
 
 11 
 
 1078 
 
 1*78 
 
 5816 
 
 6.09? 
 
 KB91: 
 
 Vivian Hurley 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 3 
 
 6 
 
 0.01? 
 
 KB92: 
 
 Children's 
 
 16 
 
 3 
 
 
 
 2 
 
 711* 
 
 
 
 10 
 
 175 
 
 920 
 
 0.96? 
 
 KB93: 
 
 Bsnrt Diablo 
 
 1 
 
 
 
 
 
 1 
 
 3 
 
 
 
 1 
 
 1 
 
 7 
 
 o.oi5 
 
 KB9 1 *: 
 
 YA Public 2 
 
 180 
 
 35 
 
 1*9 
 
 182 
 
 2579 
 
 21* 
 
 709 
 
 1*51* 
 
 1*212 
 
 i*.i*i? 
 
 KB96: 
 
 YA Public 1 
 
 181 
 
 15 
 
 52 
 
 111 
 
 2096 
 
 11* 
 
 726 
 
 381* 
 
 3579 
 
 3.75? 
 
 KB97: 
 
 West Public 1 
 
 226 
 
 23 
 
 77 
 
 161 
 
 2595 
 
 25 
 
 681* 
 
 U20 
 
 1*211 
 
 1*.1*1? 
 
 KB98: 
 
 Public 52 
 
 1*10 
 
 26 
 
 133 
 
 139 
 
 3933 
 
 18 
 
 11U6 
 
 655 
 
 6l*6o 
 
 6.76? 
 
 KB99: 
 
 Public Nl* 
 
 362 
 
 27 
 
 93 
 
 160 
 
 1*032 
 
 15 
 
 111*3 
 
 601 
 
 61.33 
 
 6.73? 
 
 KB100 
 
 Core 11 
 
 1 
 
 1 
 
 
 
 
 
 231 
 
 
 
 
 
 102 
 
 335 
 
 0.35? 
 
 Total 
 
 I 
 
 1*5!*0 
 
 1780 
 
 1061* 
 
 1895 
 
 59581 
 
 217 
 
 1261.8 
 
 13799 
 
 95521* 
 
 100.00? 
 
 Percentages 
 
 >*.75? 
 
 1.86? 
 
 l.ll* 
 
 1.98? 
 
 62.37? 
 
 0.23? 
 
 13.21*? 
 
 11*. 1*5? 
 
 100.00? 
 
 100.00? 
 
 CODE 
 
 1 = Author Search 
 
 2 = Barcode Search 
 
 3 = Dewey # Search 
 1* = Expand Command 
 
 5 = Fwd/Mext Command 
 
 6 = ICCN Search 
 
 7 = Subject Search 
 
 8 = Title Search 
 
 9 = Total Commands 
 10 = Percent 
 
 Critical Incidents. These are reports on activities or incidents of the 
 library employees that reflect favorably or unfavorably on the library. They 
 are added to the employees file after the appropriate action is taken 
 
 Item Information. The online inventory provides information on 
 each individual item in the computer. Figure 12 shows a search by subject, 
 followed by a request for the full MARC record. Also shown is a search by 
 title and author, moving forward in the file and requesting the next record. 
 
 Materials Encumbrances. The acquisition system generates a report 
 once a month that lists all encumbrances from the materials budget. This 
 report is broken down by type of item and department or division that 
 placed the order. A report is shown in figure 13. 
 
 Information and Reference Detail Report. This is a report of the 
 activities of the information services and reference division of public 
 services. Figure 14 shows the report for March 1982. 
 
PPL> S/FRONTIER AND PIONEER LIFE 
 Barcode : 1 01 1 60026 Patron : 000000000 
 
 HUNTINGTON 
 
 917-98 ON THE EDGE OP NOWHERE 
 H9510 LCCN:66-026161 
 Monument Branch 
 Subjects: Frontier and pioneer life - Alaska. 
 PPL> E/ 
 
 DueDate: None 
 
 3tUse:21-Jan-82 
 
 InvDate:13-Dec-79 
 TotC ire: 3 
 
 010 
 050 
 082 
 099 
 100 
 245 
 260 
 300 
 500 
 650 
 700 
 
 10 
 10 
 
 
 
 10 
 
 DueDate : 20-Apr-82 
 LastUse:30-Mar-82 
 
 InvDate:12-Sep-81 
 TotCirc:10 
 
 DueDate: None 
 LastUse:02-Apr-82 
 
 InvDate:11-Jul-80 
 TotCirc:7 
 
 DueDate: None 
 LastUse: 1 3-Dec-78 
 
 InvDate:30-Jan-79 
 TotCirc:0 
 
 DueDate : 28-Apr-82 
 LastUse:07-Apr-82 
 
 I nvDate : 29-May-79 
 TotCirc:13 
 
 66-026161 
 
 F909.H95 
 
 917.9803 
 
 917.98 H9510 
 
 Huntington, James, d 1915 or 16- 
 
 On the edge of nowhere, tfc by James Huntington, as told to Lawrence Elliott. 
 
 New York, 6b Crown Publishers 0c [1966] 
 
 vi, 183 p. @c 22 cm. 
 
 Autobiography . 
 
 Frontier and pioneer life 6z Alaska. 
 
 Elliott, Lawrence. 
 PPL> T/CUJO 
 
 Barcode: 1041 85327 Patron: 4001 45272 
 N KING.STEPH Checked Out 
 
 CUJO 
 
 MCN LCCN: 81 -050265 
 Penrose (Main) 
 PPL> A/BLUME.JUDY 
 
 Barcode: 10353781 7 Patron: 000000000 
 YP BLUME.JUDY 
 FOREVER 
 LCCN: 74 22850 
 Old Colo City 
 PPL> F/ 
 
 Barcode : 1 01 707404 Patron : 400000006 
 J BLUME.JUDY 
 
 DEENIE 
 
 B LCCN: 73-0801 97 
 
 Missing (PI81 ) 
 PPL> F/ 
 
 Barcode: 102434081 Patron: 400992 137 
 J BLUME.JUDY Checked Out 
 
 DEENIE 
 
 B LCCN: 73-0801 97 
 Penrose (Main) 
 PPL> N/ 
 Barcode: 1 00781871 Patron: 000000000 
 
 BLUMS* AU.L 
 
 CREATIVE DESIGN IN WALL HANGINGS 
 
 746.3 
 
 B658C LCCN : 66-0261 69 
 
 Penrose (Main) 
 
 Subjects: Hand weaving. 
 
 Wall hangings. 
 
 Textile design. 
 PPL> X/ 
 
 DueDate: None 
 LastUse :17-Feb-82 
 
 InvDate:12-Mar-79 
 TotC ire: 2 
 
 Fig. 12. Sample of the Item Information Report for the PPLD 
 
Summary Report Line Items 
 
 Leased Books :$ 7,5^6.25 
 
 Books :$UO,352.13 
 
 Microforms :$ 39.00 
 
 Subscriptions :$ 77.20 
 
 Standing Orders :$ 3, 609-57 
 
 Films & Filmstrips :$ kkk.23 
 
 Binding :$ 0.00 
 
 Recordings :$ 82.75 
 
 Framed Prints :$ 0.00 
 
 Preview Materials :$12,1*95.33 
 
 Unknown Lineitem :$ 0.00 
 
 Summary Report File Totals 
 
 Total volumes in file : 53,5^7 
 Encumbered volumes in file: 19,5^7 
 Unencumbered vol. in file : 3^,000 
 Total encumbered funds : $6^, 6h6. k6 
 Mean price per unenc. vol. :$ 3.30 
 
 Fig. 13. Materials Budget Encumbrances Report for the PPLD 
 
Figure 13 Continued 
 
 Summary Report Department & Division 
 
 RE / AD 
 
 $20 
 
 ,565.20 
 
 EX / EA 
 
 $ 5 
 
 ,110.UU 
 
 FL / MU 
 
 $ 2 
 
 ,U28.17 
 
 CH / AD 
 
 $10 
 
 ,05l.3U 
 
 LO / AD 
 
 $ 3 
 
 ,593. U8 
 
 FL / GI 
 
 $ 1 
 
 ,371.33 
 
 OU / BO 
 
 $ 
 
 51^.85 
 
 EX / BR 
 
 $ 1 
 
 ,186.15 
 
 EX / UT 
 
 $ 
 
 821. U8 
 
 EX / MO 
 
 $ 
 
 571.51+ 
 
 EX / JA 
 
 $ 
 
 813.80 
 
 AD / CO 
 
 $ 9 
 
 ,Uoi.07 
 
 EX / OL 
 
 $ 1 
 
 ,833.30 
 
 IN / AD 
 
 $ 
 
 122.30 
 
 EX / FO 
 
 $ 
 
 330.1+0 
 
 EX / PA 
 
 $ 
 
 600.89 
 
 FL / CL 
 
 $ 2 
 
 ,058.12 
 
 EX / CA 
 
 $ 
 
 760.11 
 
 OU / AD 
 
 $ 
 
 968.28 
 
 FL / TR 
 
 $ 
 
 100.00 
 
 FL / HU 
 
 $ 
 
 U30.20 
 
 FL / OT 
 
 $ 
 
 525.00 
 
 EX / AD 
 
 $ 
 
 35.1+5 
 
 AU / CA 
 
 $ 
 
 355.00 
 
 FL / PL 
 
 $ 
 
 83.00 
 
 FL / CO 
 
 $ 
 
 15.50 
 
1. Questions Answered-Directional 356 
 
 2. Questions Answered-Search-Phone 2,305 
 
 3. Questions answered-Search-Walk-in 2,lt28 
 
 TOTAL number of search questions 1*,733 
 
 k. Number of tours 
 
 5. People at tours 
 
 6 . Programs 
 
 7- People at programs 
 
 8. Outside contacts 
 
 9. People contacted outside 
 
 10. Local data base accesses 1,771 
 
 10A. Non-local data base accesses 
 
 11. Courses-Number of items added 1,^50 
 
 12. Courses-Number of items modified 15 
 13- Courses-Number of items deleted 1,3^0 
 13A. Courses-Corrections made 7 
 li*. Courses-Keywords accessed 139 
 
 15. Courses-Agencies' input hours 
 
 16. Courses-Total number of courses at the end of each month 3,059 
 
 17. Clubs-Number of Clubs added 5 
 
 18. Clubs-Number of Clubs modified 33 
 
 19. Clubs-Number of Clubs deleted 1 
 19A. Clubs-Number of corrections made 
 
 20. Clubs-Names accessed 562 
 
 21. Clubs-Keywords accessed 25^ 
 
 22. Clubs-Total number of clubs at the end of each month 58l 
 22A. Clubs-Printouts sold h 
 
 23. Call-Number of agencies added 7 
 2U. Call-Number of agencies modified ^5 
 
 25. Call-Number of agencies deleted 6 
 25A. Call-Number of corrections made 
 
 26. Call-Agencies accessed 562 
 
 27. Call-Keywords accessed 25lt 
 
 28. Call-Total number of agencies at the end of each month 68U 
 28A. Call-Printouts sold 2 
 
 29. COCIS and CIS printouts 
 
 29A. Partial Call and Calendar printouts 70 
 
 30. Community Events Calendars mailed 2^0 
 
 31. Calendar items added 139 
 
 32. Calendar items modified 9 
 32A. Calendar items corrected 7 
 33- Calendar items deleted 208 
 33A. Calendar-Keywords accessed 
 
 33B. Calendar-Total number of events at the end of each month 190 
 
 33C. Calendar-Total number of current subscribers 68 
 
 Fig. 14. Report Generated for Statistics for Information Services 
 and Reference for the PPLD 
 
DECISION SUPPOR T S YSTEM 55 
 
 Fig. 14 Continued 
 
 kk. Business Info added 
 
 52. Day Care-Agencies added 
 
 53. Day Care-Agencies deleted 
 5k. Day Care-Agencies modified 
 5^A. Day Care-Corrections made 
 
 55- Day Care-Total number of Agencies (Homes) at the end of each month 
 
 55A. Day Care-Keywords accessed 
 
 Total volunteer hours report 
 
 Periodicals Use Study. This report is a study of the use of current 
 subscriptions, and is made three times a year for a two week period to 
 provide data for subscription renewal. A partial display of this study is 
 shown in figure 15. 
 
 Challenged Material. Complaints from the public on the appropriate- 
 ness of library material are received on the form provided. Receipt of this 
 form initiates the procedure for reevaluation of material. 
 
 Interlibrary Loan Report. This report is not automated at this time, 
 but will be in the near future. 
 
 Patron Information. The circulation system stores information about 
 the persons who register. A sample of this program is shown in figure 16. 
 
 Project Status Reports. These reports are usually required by external 
 agencies. Examples are the quarterly reports required by the National 
 Endowment for the Humanities, the state library, etc. 
 
 The DSS includes utility programs to provide several types of statisti- 
 cal analysis including means, standard deviations and Pearson r (the 
 Pearson product-moment correlation coefficient r test). In addition, the 
 library has used the Statistical Package for the Social Sciences (SPSS) on 
 the University of Colorado's computer in Boulder. This program can be 
 accessed directly from one of our library terminals. A vendor-supplied 
 program for financial modeling was tried; and it was found that it was too 
 specific for our purposes and would not adapt easily to the type of statistics 
 collected and analyzed by the library. Another vendor-supplied program, 
 which calculates Beysian formulas and probability, has been received; but 
 the library has been unable to load the program into its Radio Shack Model 
 II computer. Investigation will continue for a program that will translate 
 the data or collections of data contained in the DSS automatically into 
 charts and/or graphs. One program was located that appears to do this, but 
 the license cost is $60,000. Since we believe that the cost of a program 
 
< <! 
 
 CO 5 
 
 8 o 
 
 ^ ^ 
 
 ^ 
 
 8 
 
 8 R. 
 
 p <! 
 
 ^ 5 5l 
 8~< 8 
 
 8 51 
 
 i 
 
 - 
 
 - 
 
 (U 
 
 T3 
 
 8 
 
 1 
 
 I 
 
 be 
 
 
 
DOE, JOHN G 
 
 400995887 DOE, JOHN G 
 4741 E KIOWA 
 
 Phone: 474-2255 Zip: 80909 
 Out: 000 Ovr: 000 Fines: 001.10 
 New: O8-Dec-79 Last: 06-Nov-81 
 Flags (DSCLZX): <P4 M> 
 Note: 100.01 .23-2R 
 
 Function (A,D,F,M,N,S,X): 
 
 401414933 DOE, JOHN G 
 1222 N KIOWA AV 
 
 Phone: 473-2222 Zip: 80909 
 Out: 000 Ovr: 000 Fines: . 
 New: 19-0ct-81 Last: 19-0ct-81 
 Flags (DSCLZX): <P M> 
 Note: 
 
 Function (A,D,F,M,N,3,X): 
 
 401414917 DOE, JOHN G 
 4420 E ASPEN DR 
 
 Phone: 444-2263 Zip: 80901 
 Out: 000 Ovr: 000 Fines: . 
 New: 19-0ct-81 Last: 19-0ct-81 
 Flags (DSCLZX): <P3 L M> 
 Note: RL 10.01 .81 
 
 Function (A,D,F,M,N,S,X): 
 
 401414891 DOE, JOHN G 
 2230 E KIOWA 
 
 Phone: 473-2241 Zip: 80909 
 Out: 000 Ovr: 000 Fines: . 
 New: 19-0ct-81 Last: 19-Qct-81 
 Flags (DSCLZX): <F3 L M> 
 Note: BILL 
 
 Function (A,D,F,M,N,S,X): 
 
 401414925 DOE, JOHN G 
 4420 E KIOWA AV 
 
 Phone: 473-2266 Zip: 80212 
 Out: 000 Ovr: 000 Fines: . 
 New: 19-0ct-81 Last: 19-0ct-81 
 Flags (DSCLZX): <P1F M> 
 Note: INDEBTEDNESS 
 
 Function (A,D,F,M,N,S,X): X 
 
 Circulation Department 
 
 > Menu < 
 For HELP enter "?selectlon" 
 
 EXIT CHECK GOODBYE BKSOUT 
 
 Fig. 16. Patron Information Report for the PPLD 
 
58 KEN DO WL1N b L YNN MA GRA TH 
 
 should not exceed its value to the library, we declined to purchase this 
 program. 
 
 The goal for the DSS is for the library director or the manager to use a 
 terminal to ask the DSS: How is the library today? The system would 
 respond with such comments as: "terrible," "lousy," "fair," "good," "not 
 bad," or "great." The questioner could then ask why. The system would 
 respond with a summary report of all of the indicators using predefined 
 criteria that would indicate exceptions. I believe that this is possible to do 
 for the system. The major tasks are to define the relevance of the indicators 
 and the criteria used to isolate exceptions to the norm. The Pikes Peak 
 Library District will continue to use the DSS as a management tool. It has 
 progressed far beyond the stage of just collecting numbers. 
 
 REFERENCES 
 
 1. Taggart, Marvin M., Jr., and Tharp, Marvin O. "Dimensions of Information 
 Requirements Analysis." Data Base 7(Summer 1975):8. 
 
 2. Schoderbek, Peter P., et al. Management Systems: Conceptual Considerations, rev. 
 ed. Dallas, Tex.: Business Publications, 1980, p. 8. 
 
 3. Cheney, Paul H., and Dickson, Gary W. "Organizational Characteristics and Infor- 
 mation Systems: An Exploratory Investigation. " Academy of Management Journal25(March 
 1982):181-82. 
 
 4. Anthony, Robert N. Planning and Control Systems: A Framework for Analysis. 
 Boston: Harvard University, 1965, p. 14. 
 
 5. Wagner, Gerald R. "Decision Support in the Office of the Future." Managerial 
 Planning 28(May/ June 1980):3-5. 
 
 6. Seaman, John. "Coping with the Coming Big Changes Part I." Computer Deci- 
 sions 12(Sept. 1980):76. 
 
 7. Keen, Peter, and Morton, Michael S. Decision Support Systems: An Organizational 
 Perspective. Reading, Mass.: Addison-Wesley Publishing Co., 1978, p. 1. 
 
 8. Zani, William M. "Blueprint for MIS." Harvard Business Review 48(Nov./Dec 
 1970): 100. 
 
 9. Dowlin, Kenneth E. The Promise of the Electronic Library. New York: Neil 
 S( human Press, 1982. 
 
 10. Brown, Maryann K. "Library Data, Statistics, and Information: Progress Toward 
 Comparability." Special Libraries 71(Nov. 1980):475-84. 
 
 1 1 . Library Administration and Management Association. Library Effectiveness: A State 
 of the Art. New York: LAMA, ALA, 1980. 
 
 12. Community Planning and Research Council. The Pikes Peak Library in the 1980's: 
 A Plan for Development. Colorado Springs, Colo.: CPRC, 1979. 
 
KATHLEEN M. HEIM 
 
 Assistant Professor 
 
 Graduate School of Library and Information Science 
 University of Illinois at Urbana-Champaign 
 
 Organizational Considerations Relating 
 to the Implementation and Use of 
 Management Information Systems 
 
 A management information system (MIS) is the process and structure used 
 by an organization to identify, collect, evaluate, transfer, and utilize infor- 
 mation in order to fulfill its objectives. It is a system that provides manage- 
 ment with information to make decisions, evaluate alternatives, measure 
 performance, and detect situations requiring corrective action. 1 
 
 For library managers to utilize an MIS in their operations, precise and 
 well-defined data categories are required as Runyon points out in his 
 discussion of the need for systems to assemble elusive and fugitive library 
 statistical measures. 2 Bommer and Chorba (1982) have described the use of 
 MIS for academic and special libraries in a more highly evolved mode 
 that of a decision support system with detailed consideration of manage- 
 ment reporting as a means of better identification of the activities, 
 problems and needs of users. 3 Dowlin (1980 and in these proceedings) has 
 consistently presented examples of evermore refined "up-and-running" 
 MIS in library settings with an emphasis on system components and 
 decision-making. 4 
 
 Most of the discussions addressing the use of MIS in libraries have, by 
 necessity, focused on functional areas or subsystems which affect the 
 dynamics of information and new knowledge in the following ways: 
 
 (a) stored data relations, (b) system known logical relations, (c) program 
 defined logical relations, (d) algorithm defined logical relations, and 
 (e) end-user perceived logical relations. The intent of an MIS is to pro- 
 vide the knowledge (the correct knowledge) with which to efficiently/ 
 effectively operate a system. A system can be defined as a library, a private 
 corporation, a local government, etc. 5 
 
60 KATHLEEN M. HE1M 
 
 This requisite focus on the integration of functional areas within a large 
 system in the early stages of MIS implementation fails, somewhat, to 
 consider impacts of an MIS on the organization qua organization and on 
 those who work in it. 
 
 This discussion will focus on organizational aspects of MIS imple- 
 mentation and use rather than on technical or functional issues. Because 
 few library examples of fully developed MIS exist we must turn to state-of- 
 the-art analyses of these systems' impact on organizational structure and 
 process which are lodged mainly in corporate or industrial discussions. 
 However, just as many of the principles of administrative theory formu- 
 lated for business enterprises are translatable, with modification, to the 
 nonprofit sector, so much of what is known about MIS deployment can be 
 similarly extrapolated for library considerations. In this respect we are 
 fortunate perhaps in that the slowness of libraries to recognize the manage- 
 rial implications of information generated by electronic data processing 
 systems has meant that we should be able to anticipate the problems which 
 will arise and work to circumvent them before library MIS systems are 
 more fully developed. 
 
 Organization Variables Affecting MIS Acceptance 
 
 Before we examine the effect of MIS on organizations once they are in 
 place, it is important to understand the organizational variables which 
 may affect the initial acceptance of these systems. Ein-dor and Segev (1978) 
 have characterized these variables as uncontrollable, partially controllable 
 and controlled. 6 
 
 Uncontrollable variables include organizational size, structure at the 
 time of implementation, time frame, and the extraorganizational situa- 
 tion. Larger organizations have found greater success with MIS use than 
 smaller ones. Libraries, even the largest, are relatively small organizations 
 and much planning is required to initiate and gain acceptance of MIS. 
 There will be more resistance in small organizations than in very large 
 ones since the likelihood of a lesser degree of bureaucratization and tradi- 
 tional line and staff authority divisions are in place in such operations. 
 The perception of the MIS as crystallizing these divisions may offset their 
 initial acceptance. 
 
 The more decentralized the organization at the time of MIS initiation 
 or consolidation of various components into a single system, the less likely 
 there will be a warm reception to their consolidation. This situation is an 
 important one for library planners. Independent systems for various func- 
 tions are not easy to combine, and since each is accompanied in its own 
 context by its own analysts, programmers and goals, the organization may 
 face difficulty at the time the decision to merge the systems is made. That is, 
 
MIS USE: ORGANIZATIONAL CONSIDERATIONS 61 
 
 separate acquisition, bibliographic control and circulation systems will 
 not merge easily if they are already independent entities and may compete 
 for funds. Incongruity between MIS and organizational time frames also 
 militates against acceptance of the MIS. This is a problem for production- 
 oriented organizations but may also be seen to occur in some library 
 situations as well. Generally the more relaxed the organization in terms of 
 time constraints, the easier MIS implementation will be. For example, the 
 dreaded "closing of the catalog" proclamations made throughout the 
 nation filled users and librarians with foreboding and doom. Such time- 
 frame constraints create more dissension than a more relaxed initiation of 
 systems that are more easily integrated at the organization's natural pace. 
 
 External factors such as the availability of resources for MIS imple- 
 mentation are also important. The organization with adequate data pro- 
 cessing personnel or ease of access to these people will experience greater 
 success than the organization that finds these resources difficult to mar- 
 shall. The library that must hire programmers and data entry personnel 
 from outside its own ranks will not only find difficulty in conveying its 
 needs to these "outsiders" but may not be able to sustain upgrading and 
 system maintenance. For those institutions outside of metropolitan areas it 
 will most likely be necessary to shift some personnel to permanent posts in 
 system maintenance a reallocation of resources with possible negative 
 personnel impact if done without adequate planning or anticipation. 
 
 Partially controllable variables include budgeting of organizational 
 resources, organizational maturity and the psychological climate of the 
 organization. Prior to implementation it is difficult for MIS to assume 
 imaginable cost/benefit analysis. Since they may not clearly "cost out," 
 they can only be initiated by managers with a great deal of insight. Rather 
 than mount an entire system it has been easier, especially in libraries, to 
 implement subsystems with the concomitant difficulties of consolidation 
 at the time that the full-blown MIS is desired. 
 
 Maturity of an organization is usually defined as the degree to which 
 systems are formalized, quantified and producing data appropriate to 
 decision and control. They are rational and formal. The more mature an 
 organization the more likely the MIS will be accepted since it will continue 
 the generation of these data. 
 
 The psychological climate of an organization vis-a-vis MIS is the 
 amount of expectation for the system. Most expect too much from such 
 systems at the outset, and when magical formulas for decision-making do 
 not spew forth, retreat from the system on the supposition that it has been 
 oversold. Others have heard terrible stories about MIS and bring negative 
 feelings to their implementation. The best environment for effective 
 
62 KATHLEEN M. HEIM 
 
 organizational success with an MIS is one in which preconceptions are 
 weak and realistic expectations can be developed. 
 
 Fully controllable variables include rank and location of the executive 
 and advisory committees responsible for the MIS. The more highly placed 
 the individual or committee to whom the MIS developers report, the 
 greater the likelihood of organizational acceptance. It has been found that 
 if this is more than two levels below the top of the organizational hierarchy 
 the likelihood of success is reduced. A high level steering committee to 
 guide MIS efforts, establish policy, identify potentially valuable projects, 
 and recommend resource allocations has been viewed as fundamental to its 
 acceptance. 
 
 The nature of library implementation of subsystem MIS, rather than 
 overall systems developed for general decision-making, render it difficult 
 to require that planners step back and examine the likelihood of success 
 insofar as the above factors are concerned. Given the organizational vari- 
 ables which accommodate or impede MIS acceptance and success, it seems 
 that libraries and systems with the following characteristics will find MIS 
 most compatible: those which are large; centralized; have no tight time 
 frames for the accomplishment of goals; can employ an adequate supply of 
 system personnel; have few budget problems; are already formed and 
 geared to statistics gathering and have no strong preconceptions of how an 
 MIS should be. The top executive will be fully committed to the system 
 which was planned by a steering committee and is monitored by and 
 responsible to a highly placed individual within the organization. 
 Although few libraries will fit this profile, recognition of these variables 
 may increase the capacity of system planners to understand partial failures 
 or resistance to acceptance of MIS. 
 
 MIS and Organizational Effect 
 
 The general literature of MIS and organizational effect takes two main 
 viewpoints: (1) implications for organizational structure and processes; 
 and (2) implications for managerial performance at various levels of 
 administration. For each of these we will identify aspects of special perti- 
 nence to the library and information center context. 
 
 Implications for Organizational Structure and Processes 
 
 Change in the Shape of Organizations 
 
 Although there are many ways to describe organizational structure 
 and many variants on the generally accepted pyramidal model, it is helpful 
 to imagine such a model in order to discuss current thinking on the effect of 
 MIS on traditional organizations. In such a model there are three basic 
 
MIS USE: ORGANIZATIONAL CONSIDERATIONS 63 
 
 managerial levels: top management, whose tasks are development of the 
 organization's domain, management of the interface with external envi- 
 ronments, and establishment of the organization's administrative climate; 
 middle management, which develops rules, procedures and policies in 
 order to interpret them to fit day-to-day operations; and technical manage- 
 ment, which sees that services are rendered and policies carried out. In 
 addition to this vertical model, horizontal differential may also take place 
 in varying degrees. 
 
 In libraries, horizontal specialization is nearly always functional since 
 this provides for clear task assignments and the exercise of expert technical 
 skills. A problem with this sort of horizontal development is the tendency 
 of individual units to develop their own complex communication chan- 
 nels with no gangplank mechanisms among units. This usually forces 
 coordination to the top of the organization where functional concerns 
 merge. Given this tendency, there is a natural assumption that the imple- 
 mentation of an MIS system would concentrate greater power at the 
 organization's upper strata and isolate individual units. 
 
 It is not a consideration here whether or not the way libraries are 
 organized is efficient or effective. The main question is whether or not the 
 implementation of an MIS will fundamentally alter whatever organiza- 
 tional model is in place at the time of implementation. As yet no clear 
 consensus has emerged about the impact of MIS on organizational struc- 
 ture. In their review of the literature on organizational structure and MIS, 
 Ignizio and Shannon ( 1971 ) identified two main camps: those who felt that 
 MIS would cause development of an hourglass organization with more top 
 managers, fewer middle managers and a greater ratio of skilled to unskilled 
 workers; and those who felt that the pyramid structure would bulge with 
 more management levels. Not unexpectedly, more recent observers antici- 
 pate that the MIS will become the binding force in organizations as its use 
 demonstrates the weaknesses of older structured forms and becomes the 
 model for new organizational design.* 
 
 Centralization \. Decentralization 
 
 The most hotly debated question concerning the impact of MIS on 
 organizational structure has been whether they will lead to more or less 
 centralization. Since World War II, organizations have grown more com- 
 plex and this, coupled with a human resources orientation on the part of 
 management, has meant a tendency to decentralize. However, since MIS 
 
 As one interested in the professionalization process it gave me some cheer to see the MIS 
 people struggling with the problems of more mature professions vis-a-vis their own status 
 and prestige. The exhortations of management writers that MIS technologists become inte- 
 grated into their organizations seems to be one these technicians are resisting after all, once 
 one has received holy orders is it ever possible to become one of the congregation? 
 
64 KATHLEEN M. HEIM 
 
 provide the complicated organizational communication systems required 
 to maintain control, and lessen the need to delegate authority through 
 reduction of time needed to assimilate feedback information, it is likely 
 that recentralization may occur. The reduction and regrouping of middle 
 levels of management, usually accompanying MIS implementation, also 
 contributes to recentralization. 
 
 Situational factors may be the critical component in this issue. While 
 economy of operation may be gained through concentration of informa- 
 tion at the top levels of the organization, resistance on the part of an 
 educated work force for whom creativity may be a greater satisfaction than 
 other motivational considerations, can slow this process. The initial desir- 
 ability of concentrating information at the top of an organization may thus 
 be outweighed by the need to develop a corps of managers-in-training who 
 have had decision-making experience as well as by recognition of the 
 demoralizing effects of inhibiting those at lower managerial levels from 
 exercising discretion and judgment. In an organization of professionals 
 such as a library, it is dangerous to hoard control over management 
 information. The MIS is a tool which can be used as easily for centraliza- 
 tion as for decentralization depending upon which direction the initiators 
 of the system wish the organization to move. Given the strong indications 
 of behavioral research that attest to the needs of professionals to maintain 
 autonomy and decision-making capability in order to derive satisfaction 
 from their work, it would seem unwise for any MIS system which totally 
 circumvents those whose technical expertise is needed by the organization 
 to be installed unless these individuals are seen as replaceable. It is not the 
 MIS which creates consolidation of power but those in charge of develop- 
 ing the system's use patterns. Federico (1980) has pointed out, in his 
 analysis of this issue, that the motivation and performance of middle 
 managers can suffer if the shift in control toward the top is perceived as 
 threatening the accountability and authority of middle managers. 8 
 
 Change in Control 
 
 For those top managers with an inclination to share power and 
 control of decisions, the MIS may actually make this process more palata- 
 ble. Since an MIS allows monitoring of decisions, a manager inclined to 
 share power may do so and continue to monitor the outcome with a 
 capacity to determine if a subordinate has acted in an acceptable manner. 
 Those reluctant to release authority could be apprised of the context in 
 which decisions are made at lower levels and be reassured as to their 
 appropriateness. Traditional organizational dependence on coercive 
 power can be lessened with the implementation of an MIS since, theoreti- 
 cally, information could be shared by those in the best position to use it. 
 Argyris (1982) has hypothesized the development of a project team or 
 
MIS USE: ORGANIZATIONAL CONSIDERATIONS 65 
 
 matrix form of organization designed around relevant information rather 
 than formal power. 9 Peer relationship might be emphasized and members 
 of matrix groups assigned equal power to make decisions with the ultimate 
 result a lessening of superior/subordinate identification. The potential for 
 participatory management and maturity of the individuals in an organiza- 
 tion could then be enhanced. 
 
 Such a development, however, will call for a new breed of executive 
 officer at the top of an organization. If competence and technical knowl- 
 edge replace formal authority, those top managers who maintain control 
 by withholding information will find themselves less powerful and less 
 essential. The MIS, if used openly, could cause a major organizational 
 restructuring from arbitrariness to explicitness. 
 
 The three impacts of MIS on organizational structure so far discussed: 
 (1) change in shape of the organization, (2) centralization v. decentraliza- 
 tion, and (3) change in the control and authority structure of the organiza- 
 tion do not happen in isolation. As we have already pointed out the 
 attitude of top management toward the use of MIS may alter the direction 
 taken vis-a-vis these three impacts. 
 
 Alteration of Organizational Processes 
 
 MIS alter organizational processes as well as organizational structure. 
 The large general literature on "change" and change agents is helpful in 
 understanding means to mitigate the effects of MIS installation. Foresight 
 into potential organizational transformation is difficult, but Federico 
 discusses axioms of alteration such as those developed by Benne and 
 Birnbaum to mitigate the negative affects of MIS implementation. These 
 include: 
 
 1. changing of all relevant aspects of the system (not just obvious 
 ones); 
 
 2. complementary and reinforcing changes on all levels; 
 
 3. introduction of change at stress points since these are the areas 
 most amenable to modification; 
 
 4. consideration of informal as well as formal organizations; and 
 
 5. inclusion of those affected in the planning of the change. 10 
 
 Organizational acumen is a key factor in MIS implementation. Anticipa- 
 tion of the variety of organizational impacts and action to develop balances 
 in the new system are crucial. 
 
 Change in organizational processes introduced may be examined at 
 both the individual and general personnel levels. At the individual level, 
 Coleman and Riley (1972) have noted that change caused by MIS creates 
 conflict and stress which generates resistance arising from inaccurate 
 perceptions of the effect of the MIS on the organization fear of the 
 
66 KATHLEEN M. HEIM 
 
 unknown; anxiety arising from enlarged responsibility; threat to position 
 and stature; and disruption of personal relations. 11 Petroff ( 1973) has added 
 the fear of more precise personnel evaluation. 12 
 
 The literature concerning the effects of MIS on general personnel 
 issues includes discussion of changes in job content, task requirements and 
 retraining. While upper management may experience broadening and 
 increased responsibility, jobs lower in the organization may become more 
 routine and reduced in content. In fact, those higher in the organization 
 have indicated more satisfaction with MIS than those lower. 13 At clerical 
 and supervisory levels, interpersonal relation skills become less important 
 while at the higher levels more openness is required. These effects may find 
 the organization with a highly efficient MIS to have two different adminis- 
 trative philosophies in operation: a near return to Taylorism at the lower 
 levels and an extreme case of human resources orientation at the higher 
 levels. Since these styles tend to be antithetical, very real personnel prob- 
 lems may develop. A number of MIS analysts agree that personnel prob- 
 lems associated with MIS implementation cause more disappointments 
 and failures than technical problems. In an organization composed of a 
 great many professionals, the need for creativity and individuality are 
 high. Insofar as an MIS may concentrate these job requirements at a few 
 levels, important motivators may be removed from the workplace. 
 
 The threat of depersonalization and personnel perceptions that their 
 positions have been reduced to MIS created niches reflecting only the 
 needs of the system rather than the employee, rank high in the list of 
 problems which face the organization attempting to implement an MIS. 
 Libraries are particularly susceptible to this problem since low salaries 
 mandate that those employed derive greater nonhygienic benefits in order 
 to sustain motivation. Removal of motivational factors, most libraries' 
 only means of providing job satisfaction, may undermine the rationale for 
 staying on the job. Finally, one rather short-term effect of the MIS on 
 organizational processes must be noted. In the introductory stages an MIS 
 will cause lags in the organization's progress. Routines that were relatively 
 efficient, or seemed to be so, will be disrupted as new ones are introduced. 14 
 While this effect should be overcome as the system becomes familiar to 
 employees, short-term major disruptions will generate ill will toward the 
 new system unless they have been well prepared for. 
 
 The three main impacts of MIS on organizational processes are: 
 (1) radical change, (2) individual and general personnel dissatisfaction as 
 job content is revised and new task clusters evolve, and (3) disruption of 
 routine. If anticipated and planned for, these can be minimized, but if MIS 
 are introduced without adequate attention to these factors the system may 
 have difficulty due to personnel resistance. 
 
MIS USE: ORGANIZATIONAL CONSIDERATIONS 67 
 
 Implications of MIS for Management 
 
 The most salient question regarding MIS and its effect on manage- 
 ment is whether an MIS will fundamentally alter managerial functions as 
 they are practiced. If an erosion of traditional management responsibilities 
 occurs with MIS implementation, the general conception of what com- 
 prises management skills may well change. Most researchers agree that the 
 effects of MIS will be very different on different levels of management. At 
 the highest levels, managers with the support of an MIS should be able to 
 focus more intensely on innovation and change, develop alternative simu- 
 lations for problems to be solved, investigate up-to-date research findings, 
 avoid routine decision-making, and shunt organizational loyalty consider- 
 ations in favor of more rational concerns with difficult problems. 15 Middle 
 managers, on the other hand, could find their work more highly structured 
 and reduced in status. The truncation of the middle manager role would 
 require more specialization and less scope of action. We have already seen 
 in our discussion of the effects of the MIS on the structure of the organiza- 
 tion that the role of the middle manager could go either way. 
 
 This potential change in the practice of management is ironic in the 
 face of recent investigations of the styles of administration in Japanese 
 firms with their focus on maximization of human resources. Since a central 
 factor in the success of these firms is in their commitment to the consulta- 
 tive style of decision-making, structured to involve the whole group rather 
 than a few individuals, the implementation of MIS in terms of organiza- 
 tional behavior and shifts in managerial style may move us farther from the 
 successful modes of management and back into an earlier more centralized 
 phase. 
 
 Elsewhere in these proceedings, Olsgaard addresses factors involved in 
 top management's use or nonuse of information so we will comment here 
 primarily on effects of an MIS at lower levels in the organizational 
 hierarchy. 
 
 The horizontal effects of MIS implementation are of special impor- 
 tance when trying to assess the role of the MIS at lower levels of manage- 
 ment. The MIS as an integrated computer-based system for providing 
 information to support operations and decision-making tends to be quite 
 useful at middle-management levels if in fact a more flexible view of an 
 MIS is understood in its operational mode. 
 
 The more information available to a manager, the more involved 
 she/he will be in her/his commitment to the organization's goals. The 
 traditional functional division of library operations without solid interde- 
 partmental communication gangplanks can cause isolation and power 
 hoarding in individual units. Since an integrated MIS could conceivably 
 open a system and enable qualified users to peruse various aspects of the 
 
68 KATHLEEN M. HEIM 
 
 operation tangential to the designated area of concern, the context of 
 individual decisions might be made with a better understanding of where 
 the organization is at any given point. An MIS will facilitate horizontal 
 communication since it will force consistent definitions and formats. 
 Interdependence of units should increase. 
 
 In a library setting all this is speculation since, in a service organiza- 
 tion with rather pure missions and goals, considerations of unit power and 
 control should be moot. Ideally there would be no need to wrest power or 
 importance for a particular unit since goals should be kept in mind more 
 consistently through such devices as the planning process for public 
 libraries or frequent analyses of objectives via MRAP (Management 
 Review and Analysis Program) 1 in academic settings. 
 
 However, the rise of MIS has seen a shift, perhaps a short-term one, but 
 nevertheless a shift, in perceptions of unit power in larger libraries. Pres- 
 tige and status accrue to those who work in departments with greater MIS 
 capabilities. The old technical/public services dichotomy, with the 
 implicit emphasis on public service, has blurred and the action, the excite- 
 ment, the pioneering edge of librarianship seems now to be the realm of 
 technical services. The increase of public service literature focusing on 
 online searching or computer-assisted instruction (CAI) underscores this 
 observation. The cachet which comes with synergistic innovation with a 
 CRT seems to add prestige to those who work plugged into an electric 
 keyboard. For the time being the technical services' development of MIS, 
 both locally and through networking, is far ahead of those in the public 
 services. These events may create a short term imbalance of departmental 
 power and tempt units into competition a dysfunctional situation for the 
 organization's mission. 
 
 Saunders's examination of MIS and departmental power has some 
 applicability to libraries. She defines power as the capability of a subunit 
 through formal position or actual or perceived participation in organiza- 
 tional activities to exert influence on another subunit to act in a prescribed 
 manner. 17 If subunits vie for scarce organizational resources, especially 
 personnel, there must be mutual assessment of power bases. The ability of 
 one department to exert influence on another is determined to the extent to 
 which it participates in organizational decision-making on key organiza- 
 tional issues. These may change over time or be different for any given 
 institution. An ARL library with its concentration on collection develop- 
 ment and maintenance may find that the bibliographic units are more 
 important than the public service units, especially if the administration is 
 more committed to number of volumes and depth of collection than 
 service. A library serving a research and development function, on the 
 other hand, may be so committed to the support of research that the public 
 service function takes precedence. 
 
MIS USE: ORGANIZATIONAL CONSIDERATIONS 69 
 
 The use of an MIS may affect the power of units in three ways: (1) in- 
 creased access to information may allow subunits greater influence in 
 organizational decision-making on key issues; (2) the capacity to cope with 
 uncertainty may also grant greater power within the organization to the 
 degree to which the unpredictability ensuing from lack of information 
 about future events may be reduced; and (3) the nonsubstitutability or 
 difficulty with which the activities of a unit may be performed by an alter- 
 nate department. These affect pervasiveness and number of linkages with 
 other units. 
 
 Task criticalness and the degree to which the activities of a unit affect 
 the achievement of the main goals of the organization is a mediating 
 variable. 18 Depending upon the library's long term goals, task criticalness 
 may shift and create deceptive short-term power imbalances within units. 
 A good example is closure of the catalog. While one of the ultimate goals is 
 provision of multiple access points, achievement of that goal has involved 
 a series of changing power bases within the library. Hardware developers, 
 software programmers, catalogers, and ultimately public service librarians 
 have all participated in effective use of this tool. As tasks critical to the 
 goal's success have changed, so has the relative power of units associated 
 with each step. While this shift has short-term disruptive effects, the 
 long-term goal will be met and, insofar as units subscribe to the organiza- 
 tion's mission, competition avoided. Those responsible middle managers 
 who experience shifts in power as various subunits rise and fall in their 
 power base, must be ready to accept the changing perceived measure of 
 status. 
 
 Since MIS can enhance the power of organizational units, another 
 administration consideration must be how important power may be to 
 managers. Job satisfaction studies which have demonstrated a positive 
 correlation between employer performance and perceived status of the 
 manager and the power of the unit should be considered in terms of 
 changing unit dynamics due to the better availability of information. An 
 MIS capacity to generate too much information, alluded to in the keynote 
 speech, is also a determinant of employee satisfaction. O'Reilly (1980) has 
 shown that perceptions of over- or underload correlate with satisfaction 
 depending upon the manager's style. 19 
 
 For the middle manager attuned to organizational goals, the effect of 
 MIS can be quite positive. New communication patterns can be estab- 
 lished, better decisions can be made, and more shared knowledge can be 
 available. These factors could prove disruptive to the empire builder, 
 however, since, in the long range, MIS should function to streamline the 
 overall organization to the detriment of unit power, although short-term 
 and somewhat misleading expansion of unit power may take place. From 
 the larger organizational perspective, this evolutionary and dynamic 
 
70 KATHLEEN M. HEIM 
 
 nature of MIS's effect on middle management should be anticipated and 
 planned for. 
 
 Conclusion 
 
 The potential of MIS for better organizational decision-making and 
 better deployment of organizational resources is great. However, in librar- 
 ies this capacity has generally been discussed without adequate attention to 
 the complex factors of organizational structure and processes or the resul- 
 tant effects on individuals; restructuring of institutional personnel poli- 
 cies; resultant shifts on the demands of top and middle management; or 
 changes in unit to unit communication, power and satisfaction. 
 
 The voluminous literature on these aspects of MIS implementation in 
 the general management literature are inconclusive and confusing. On 
 either side of any issue a number of studies support the direction in which 
 each of these organizational factors might move. However, as with any new 
 innovation, the addition of MIS capacities to organizational life requires 
 careful consideration of the human element in individual and group 
 interaction. The lure of precision through information, economy through 
 better understanding of quantifiable variables and efficiency through 
 clearer analysis of service and production may so alter the organization 
 that those in charge of the organization's direction may find its workers 
 (both professional and support staff) confused, less satisfied and alienated. 
 
 Sterling (1980) has observed that MIS systems and their concentration 
 of feasibility, workability and minimization of costs have failed to focus 
 management concern on the antihuman aspects of such efforts. 20 Since 
 automation of any management system codifies the rigidity of practice and 
 expands it to ever larger circles, the prerogative to formulate questions 
 important to the human element of the organization is diminished. In 
 conclusion, I would like to caution that the glamour of MIS and their 
 capacity for variant simulations of organizational outcomes be considered 
 carefully in light of the human factors in an organization. The paucity of 
 service organization studies makes their advancement into MIS implemen- 
 tation even more uncertain than in those that are profit based. We simply 
 do not know what will happen but we must recognize that the human 
 factor has played a role of great importance in libraries to date and not 
 forego our investment in the development of a highly skilled and techni- 
 cally competent corps of professionals in favor of efficiency and streamlin- 
 ing of operations. 
 
 At this critical stage of MIS innovation, with seemingly unlimited 
 technological opportunities, it is more important than ever before that the 
 organizational and individual consequences be attended to, analyzed, and 
 considered as major managerial adjustments involving MIS are 
 undertaken. 
 
MIS USE: ORGANIZATIONAL CONSIDERATIONS 71 
 
 REFERENCES 
 
 1. Morss, Elliott R., and Rich, Robert F. Government Information Management: A 
 Counter- Report of Commission of Federal Paperwork. Boulder, Colo.: Westview Press, 1980, 
 p. 137. 
 
 2. Robey, Daniel. "User Attitudes and Management Information System Use." 
 Academy of Management Journal 22(Sept. 1979):527-38; and Runyon, Robert S. "Towards 
 the Development of a Library Management Information System." College & Research Librar- 
 ies 42(Nov. 1981):539-48. 
 
 3. Bommer, Michael R.W., and Chorba, Ronald W. Decision Making for Library 
 Management. White Plains, N.Y.: Knowledge Industry Publications, 1982; and Bommer, 
 Michael R.W., etal. "Performance Assessment Model for Academic Libraries. " Journal oj the 
 ASIS 30(March 1979):93-99. 
 
 4. Dowlin, Kenneth E. "A Public Library Management System." In Library 
 Effectiveness: A State of the Art, pp. 85-110. Chicago: ALA, 1980. 
 
 5. Mendiville to Heim, personal communication, 5 April 1982. 
 
 6. Ein-dor, Phillip, and Segev, Eli. "Organizational Context and the Success of 
 Management Information Systems." Management Science 24(June 1978): 1064-77. 
 
 7. Ignizio, J.P., and Shannon, R.E. "Organization Structures in the \98Q's." Industrial 
 Engineering 3(Sept. 1971):46-50. 
 
 8. Federico, Pat-Anthony, et al. Management Information Systems and Organizational 
 Behavior. New York: Praeger, 1980. 
 
 9. Argyris, Chris. "Organizational Learning and Management Information Systems." 
 DataBase 13( Winter/Spring 1982):3-11. 
 
 10. Federico, et al., Management Information Systems, p. 104. 
 
 11. Coleman, R.J., and Riley, M.J. "The Organizational Impact of MIS." Journal of 
 Systems Management 23(March 1972): 13- 19. 
 
 12. Petroff, J.N. "Why are DP Managers So Unpopular?" Datamation 19(Feb. 
 1973):77-79. 
 
 13. Federico, et al. Management Information Systems, pp. 110-18. 
 
 14. Ibid., p. 160. 
 
 15. Ibid., p. 38. 
 
 16. For a review of MRAP see: Johnson, Edward R., and Mann, Stuart H. Organization 
 Dei'elopment for Academic Libraries: An Evaluation of the Management Review and Analy- 
 sis Program. Westport, Conn.: Greenwood Press, 1980. 
 
 17. Saunders, Carol S. "Management Information Systems, Communications, and 
 Departmental Power: An Integrative Model." Academy of Management Review 6(July 
 1981):433. 
 
 18. Ibid., p. 437. 
 
 19. O'Reilly, Charles A., III. "Individuals and Information Overload in Organizations: 
 Is More Necessarily Better?" Academy of Management Journal 23(Dec. 1980):684-96. 
 
 20. Sterling, Theodore D. "Humanizing Computerized Information Systems." In 
 Management System in the Human Services, edited by Murray L. Gruber, pp. 287-99. 
 Philadelphia, Pa.: Temple University Press. 
 
WAYNE MULLIN 
 
 Head Loan Librarian 
 University of Arizona Library 
 
 Geac as a Source of 
 Management Information 
 
 When Professor Lancaster called to ask if I would be willing to give a paper 
 on management aspects of the Geac Online Circulation System,* I was 
 both elated and apprehensive because the University of Arizona was in the 
 unique position of just having gone from a nonstandard version of Geac 
 software to the standard 4.0 turnkey version. At the time I accepted the 
 invitation, no one at the University of Arizona knew much about what 
 management data Geac could provide. I am obliged to tell you straight- 
 away that there is still much that we at Arizona do not know about the 
 management data Geac can provide. But in the almost two years we have 
 been on Geac, we have learned a thing or two some of which relates to the 
 topic of this Clinic. 
 
 Among the more interesting things learned was that Geac came in 
 several "flavors," at least from June 1980 through June 1981. At installa- 
 tion and through the first year, ours was not "vanilla" a pure turnkey 
 system but rather what came to be dubbed by us as "marble fudge." We 
 think that we are the only marble fudge system in existence, but Geac 
 would be a better source of this information than I. What I am saying is 
 that our Geac 8000 looked, from the outside, like a turnkey Geac; per- 
 formed somewhat like a turnkey Geac; but, oh boy, it sure didn't work and 
 look like a turnkey Geac when one got inside the guts of the thing. The 
 upshot of being marble fudge was that we did not have many reports until 
 installation of the version 4.0 thirteen months after coming live. 
 
 This paper describes the Geac Online Circulation System as it existed in March 1982. 
 Readers should be aware that there have been numerous changes and improvements during 
 the past year. 
 
 72 
 
GEAC 73 
 
 To fully appreciate or at least to appreciate to some degree what I 
 will say this afternoon is to have some understanding of my background 
 and the environment at the University of Arizona Library. First, I am a 
 librarian. One of those people who got turned on to libraries from having 
 worked in one as a support staff and I know well that some of you can talk 
 circles around me when it comes to understanding how computers really 
 work, or sometimes do not work at all. What I will cover then first is a 
 librarian's view of the Geac system not the computer specialist's perspec- 
 tive. I think this is fair since Geac promotes itself as a system for the 
 uninitiated user. Second, note that I am not a Geac salesperson, but, on the 
 other hand, I am not a severe Geac critic either. I suspect, though, that I 
 will seem a little of each. Third, the University of Arizona Library does not 
 have trained computer and systems specialists among its staff. Our current 
 head computer operator is an English major who luckily happened to be 
 employed in the loan department of the university at the time we consid- 
 ered online circulation systems. I say luckily because he (Tom Owens) had 
 worked at Ohio State at the time they were developing their circulation 
 system. 
 
 By way of further introduction, I think it will be interesting to note 
 what the University of Arizona asked for (and did not ask for) in its Request 
 for Proposal (RFP) for an online circulation system under the general 
 heading: Management Reports/Notices/Lists. That this request took up 
 most of what was to become page thirteen of the RFP should have given me 
 a clue to our future, both from the standpoint of not asking for the obvious 
 and of asking for the impossible. In quick summary, we wanted: 
 
 1 . Circulation activity as follows: 
 
 a. patron category, 
 
 b. call number, 
 
 c. classification number, 
 
 d. library, 
 
 e. call number linked to patron category; 
 
 2. lists of missing items/replacement items; 
 
 3. lists of items that circulate for more than a specified number of times as 
 determined by the library; 
 
 4. lists of items that have not circulated in a specified period of time as 
 determined by the library; 
 
 5. lists of patrons who are delinquent, owe fines, etc.; 
 
 6. overdue search lists; 
 
 7. lists of items with more than a specified number of holds or recalls as 
 specified by the library; 
 
 8. a daily circulation list on microfiche; and 
 
 9. a hold shelf clearance list with cumulative statistical capabilities. 
 
74 WAYNE MULLIN 
 
 The impossible to secure was circulation information by call number 
 linked to patron category. It was an important piece of information in that 
 we had just gone through the exercise of trying to demonstrate what items 
 faculty and graduate students used in order to comply with U.S. Govern- 
 ment standards for the allocation of research overhead. Research overhead 
 to the University of Arizona Library represents a considerable chunk of 
 money each year. 
 
 Not asking for the obvious was specifically not requiring that Geac be 
 able to tabulate circulation totals for a given period of time as a standard 
 report. To my amazement, the system still cannot give total circulation 
 counts as a standard report. Why? Because the TSTA Report does not 
 count renewals. Our head computer operator, Tom Owens (who assisted 
 greatly in the preparation of this paper), has written a program to get 
 circulation totals for periods longer than one month (usually for a semes- 
 ter). This report does not count renewals either. The history file must be 
 used to do this. Tom's program requires asking for how many items have 
 circulated more than zero times in a specific location and then cross- 
 multiplying and adding totals for each location in the library system. I tell 
 Tom that he is running the most expensive manual calculator in the 
 history of humankind, but he says research may prove me wrong. 
 
 In this discussion we will look at four general areas of the Geac system: 
 what I call circulation management, financial management, the biblio- 
 graphic and patron extract modules, and finally, system management 
 how the Geac manages itself. I will begin with circulation. 
 
 One of the most often asked questions I hear is: How often did a 
 particular book circulate? Geac provides such information online at the 
 copy level of the bibliographic record. By finding a specific title and then 
 going to the item level screen and keying a copy detail command, one can 
 tell how often a particular item has circulated up to a maximum of 256 
 circulations. Libraries having heavy circulation, such as for reserve mate- 
 rial, are limited by the 256 count (this will come up again when I talk about 
 reserve circulation). These counts must be zeroed out each semester so that 
 we can obtain accurate reserve circulation counts. Other information 
 available in the online copy detail display is the date the record has last 
 been edited, the date the system was entered, and the number of times an 
 item has been reshelved. 
 
 More general, but important, circulation information is available in 
 Geac reports. One of the reports the University of Arizona has been waiting 
 for and just recently received is the Interagency Activity Report. In essence 
 this report gives a general profile of who is using what kind of materials. 
 By defining item material types with some specificity, a library will be able 
 to profile general collection use. 
 
GEAC 75 
 
 The report is sorted by agency, patron privilege type and material 
 type. Patron privilege type is part of the matrix controlling the loan 
 period. It is more general than patron statistical type which can identify 
 users within narrow categories and is counted in TSTA. Material type is 
 also part of the loan matrix. 
 
 In looking at the report, one can see that the privilege type is repeated 
 for each material type on which activity appears, but we plan to change 
 this display to show privilege type only once. Activity can include circula- 
 tion (charge out) information as well as information on discharges, books 
 placed on reserve (RRP), fines action, holds placed, and partial payments 
 made in the financial module. 
 
 The report totals the activity areas within privilege class and material 
 type. It also totals by specific activity and agency. For libraries using 
 departmental codes, such as for academic departments in a college or 
 university, the report also gives total activity. Agency totals are also shown. 
 
 After loading departmental codes, we realized that the required man- 
 ual input of patron data would be too complex and time consuming. We 
 had entered over 100 codes. Thus, all users at the University of Arizona 
 default to one "department." The report cumulates to the point when the 
 file is so large that it needs to be dumped onto tape, for example. The file 
 then zeros out and a new cumulation begins. 
 
 One of the things to remember when I discuss various reports is that 
 local policy, database size and long hours of operation can hamper full 
 utilization of some Geac reports. The University of Arizona's Geac data- 
 base contains over 1.6 million item records represented by approximately 
 1.2 million titles. There are over 60,000 users in the pa iron database. Geac's 
 report that lists all of the items' statistics is an example of a report that 
 cannot be successfully run in a large library at least not at the University 
 of Arizona. 
 
 The report produces a list of all item statistics item being the specific 
 copy of a work. The report is sorted by agency, branch (terminal location), 
 and transaction type. The transactions one might want to report on 
 include charges/discharges, RRPs, fine information, holds, partial pay- 
 ments, or all of these areas. 
 
 The sample report from Geac was sorted by charges/discharges. Infor- 
 mation reported was the call number, item type, item number, item loca- 
 tion, terminal number, transaction date, transaction time, patron number, 
 and patron type. As one can see, for a particular agency and specific date, a 
 wealth of information is available. Finding time to use it all, however, is 
 another matter. 
 
 One problem most libraries face is what to do about missing items. 
 The management of missing books has been handled by Geac in two ways: 
 
76 WAYNE MULLIN 
 
 through the Missing Report and through the utilization of the system's 
 local programming options using a Geac language called GLUG. I want 
 to take a minute and talk about this powerful language with such a funny 
 name. 
 
 GLUG is the report-writer language for the Geac system. An inter- 
 preted language, it allows the user, even one with moderate sophistication, 
 to create reports, secure in the knowledge that it is almost impossible to 
 make a mistake which would damage the database or affect software 
 support maintenance from Geac. The driver program that interprets the 
 GLUG code has been written so that no data files may be modified by a 
 GLUG program. 
 
 Although the language has limitations, it surely would meet the needs 
 of any but the most sophisticated computer operators. GLUG is particu- 
 larly useful because it can easily overcome the limitations of the biblio- 
 graphic and patron extract programs which will be presented later. 
 
 Staff at the University of Arizona were able to develop useful GLUG 
 programs about two months after the language was made available to us. 
 As an example, our head computer operator has had no formal training in 
 programming and very little experience (as mentioned earlier he some- 
 what slid into his present position when the system was implemented). He 
 has had no problems developing GLUG programs despite his lack of 
 training. We believe most libraries will find the language easy to use 
 regardless of the level of staffing. Combined with the extract programs, the 
 GLUG language furnishes a package of programs which will allow a 
 library the ability to create many of the reports it needs. 
 
 To get back to missing books, the areas in which the University of 
 Arizona needed much help with was good management of what items were 
 missing, no longer missing or still missing at the end of one year. Geac 
 produces, by operator action, a report of missing items. It is sorted by 
 agency, the missing status (there can be a variety of reasons why an item is 
 missing), and by call number. We at Arizona are most interested in the 
 status information. An item listed as missing but subsequently checked out 
 would show as "returned from missing" on this report which modifies the 
 online record to show that the item is no longer missing. The report was 
 particularly useful in the transition from the old missing procedure to the 
 one that is now automated. The zero indication in the status field of the 
 report told us to pull the manual missing card for appropriate action. 
 
 Under new procedures, the Bibliographic Extract and a Geac/GLUG 
 program, modified locally by the head computer operator, is used to 
 identify those items missing in the system one year after being declared 
 missing. One of the optional bibliographic fields is used to input the 
 month and year the item is declared missing. One year later the Biblio- 
 graphic Extract program is run, asking for all books still missing from 
 
GEAC 77 
 
 March 1981, for example. The report is sorted by call number which 
 enables staff to search for the material and is printed by the locally 
 produced program. Those items found are discharged, which "frees" the 
 missing status when the Geac Missing Report is run. The Bibliographic 
 Extract program is then run again, dropping all the discharged items. 
 What remains are those items missing for one year or more. This informa- 
 tion is given to the catalog department so that appropriate cards can be 
 withdrawn from the various catalogs, and bibliographic and/or item 
 information deleted from the Geac database. 
 
 An additional concern for many libraries, depending upon lending 
 and other policies, is that of the user whose card has expired in the system 
 but who still has material checked out. Geac addresses this management 
 situation through a program that produces a list of all card expired patrons 
 with items checked out. 
 
 The report is sorted by agency and then alphabetically by patron 
 name. Patron information provided besides name includes: address, phone 
 number, ID number, and expiry date in the system. Each separate item 
 checked out is listed after the patron information. Thus, if a user had ten 
 items checked out, ten complete patron entries as described earlier would 
 appear. Item information includes call number, author, title, material 
 type, due date, and status other than normal (such as billed for replacement 
 or missing). The report could be useful for a small library which needed to 
 keep track of such information. It could also be useful in any library whose 
 loan policy prevented loan periods to exceed the expiry date. A library 
 could influence when a person showed up on this report (at least in theory) 
 by staggering patron expiry dates in the system. 
 
 At the University of Arizona this report is of little use because we do 
 allow loan periods to exceed expiry date. One of the reasons we allow this is 
 that library staff could not handle successfully the large end-of-term 
 and/or academic year return or renewal of all materials. Thus, this poten- 
 tially excellent report is not often run. 
 
 It would be unwise to move on to other areas without talking about 
 hold management information and class reserve capabilities. Geac has a 
 variety of report programs dealing with holds. A hold is defined here as 
 that process by which a user can be notified that a specific item has been 
 returned and is waiting on the hold shelf to be picked up. 
 
 One of the things called for in the University of Arizona's RFP was a 
 Hold Shelf Clearance Report. My thought was to identify those items 
 which had not been picked up so that they could be pulled from the shelf, 
 and also to keep statistics on the number of such items. (I must tell you that 
 I was alone in wanting such a report. Staff in the hold unit saw little use for 
 it.) 
 
78 WAYNE MULLIN 
 
 What we received was a report that did list such items, but that also 
 listed holds which had been terminated. (There are at least two ways a hold 
 can be terminated in the system. One way is for it to have been "sitting" for 
 180 days waiting to be acted upon when the item is returned. The other is 
 for a staff member to go into the system and terminate the hold.) This list, 
 then, gave us much more than we needed to do the job; so much more that it 
 was far easier to go through the hold shelf manually looking for items that 
 needed to be pulled. However, because the Hold Report was in theGLUG 
 language, we were able to modify it locally without fear of losing software 
 support for the system itself. As a point of information, Geac was unable to 
 provide statistical cumulations in this report. 
 
 The report is not useless, it is just that we feel it cannot be used as 
 intended and I emphasize the we. The report could be used in libraries 
 that need more control over, or are interested in, more detailed statistical 
 information about holds. It does give one a good idea of items not picked 
 up and the number of items lapsed in the system. By working with the 
 provided information one could: (1) keep track of which patrons are not 
 picking up holds in order to see if there are notification or other problems 
 in the hold routine; and (2) keep track of the number of holds terminated 
 (not satisfied) by the system. 
 
 Three other hold program reports need mentioning. One of these is a 
 report that produces a list of items on which holds have been placed 
 between certain dates determined by the library each time the report is run. 
 Also identified is the patron who requested the hold. It is sorted by the 
 pickup agency and then arranged alphabetically by patron name. When 
 we first looked at this report, we did not quite know how it could be used. 
 Geac, I suspect, like other turnkey vendors, does give descriptions of 
 reports, but does not tell what it might have had in mind in designing the 
 thing. Some reports are rather obvious in their utilization. The Returned 
 From Missing Report needs little context beyond its description. 
 
 After talking about this particular report, it did seem that we could 
 have used it if we had so desired. By going into the system and seeing if a 
 recall had also been placed on the item shown, we could, through the 
 timing of the report run, determine what items were not returned by the 
 adjusted recall due date. (It is important to note that one should place a 
 hold on an item when recalling it, although the system will allow a recall 
 without identifying the person needing the material.) 
 
 One might think that this report could help measure work load. That 
 is, by running the report daily or weekly, for example, one would know the 
 number of holds placed. This is true, except that the moment a library 
 places terminals for use in public areas and allows users to place holds 
 themselves, the work load utilization is lost. Other libraries may indeed 
 
GEAC 79 
 
 find the report useful, but based upon policy and procedures in place at the 
 University of Arizona, it is not used. 
 
 Geac also provides a program which lists the contents of the hold 
 shelf. It is sorted by patron name within pickup agency and lists the patron 
 name, phone number, item information, active and expiry dates. One 
 could also use this list to notify patrons of available material. Arizona does 
 not use this report, but relies instead on the availability notice that is 
 generated when a book with a hold on it is discharged. 
 
 The most heavily used holds report at the University of Arizona is the 
 Holds/ Purchase Alert Report. It works in part within parameters set by the 
 library in that it allows a library to set the ratio against which the system 
 searches for hits. We use the report to alert us to the possibility of ordering 
 additional copies of high demand material. Currently, three holds against 
 a single copy will trigger inclusion in this report. The bibliographic 
 information is then sent to the acquisitions department for a decision on 
 ordering additional copies. 
 
 Since we have just begun to work with this report, it is too early to 
 measure its utility in our particular setting. The report is sorted by location 
 and then by call number. Information displayed includes call number, 
 author and title, location of item, active copies, missing copies, total 
 copies, patron name and phone number, and specific number of the copy 
 on which the hold has been placed. Active copies are those that are 
 displayed when looking at item information, as opposed to copies which 
 have been deleted. Total copies include a running total of all copies that 
 have ever been in the system. 
 
 The Data Entry Report, also known at the University of Arizona 
 Library as the infamous accession list, comes in two standard formats: long 
 and short. If one is familiar with the full bibliographic screen in Geac, one 
 gets a good idea of information contained on the long list. If not, it 
 includes: call number, title, author, note, status, entry date, last edit date, 
 copies, accession number, price, barcode number, copy status, copy refer- 
 ence number, user copy number, material type, circulation information, 
 reshelve information, last use date, and all the optional fields. The report is 
 sorted by location and operators' initials. The short form includes call 
 number, title, author, copies, volume, material type, location, date the 
 record was created/ updated, and operators' intials. 
 
 The Data Entry Report is an excellent source for checking the accu- 
 racy of input to the system. In trying to use the short form as an accessions 
 list, however, it fails badly in a number of ways. I mention its use as an 
 accessions list because this is what the Geac system said would fulfill the 
 RFP. Some of the failures are no fault of Geac. We simply do not agree on 
 the definition of accessions which we view as new items not retrospective 
 
80 WAYNE MULLIN 
 
 conversion items. What are the problems with the Data Entry Report? 
 First, the call number field of the report is not large enough to take a long 
 call number. Thus, using the list to identify new material and then to 
 locate the item in a library is often difficult or impossible. Second, the 
 author and title fields are often not large enough to display sufficient 
 information. The field sizes might also be an argument against this 
 intended use, except that we believe the full report does give complete 
 information. 
 
 The real problem in using the short form as an accessions list is that it 
 shows every record entered into the system from any source. Thus, it is 
 much more than a new accessions list. Hence the name: Data Entry Report. 
 It is so logical when one thinks about it, but it surely doesn't work for us. 
 
 We have reworked the report through a locally written program to 
 eliminate extraneous information and expand the call number field so that 
 a full number is displayed. This can all be done without altering the Geac 
 software supporting the standard report. To get around the too much 
 information problem, we are working on a program that will hit against a 
 combination of OCLC number or one other hook such as LCCN or ISSN 
 numbers. Our head catalog librarian feels that the retrospective input now 
 being done in branch collections will not often have an OCLC number, 
 but may have other hooks. All data being entered through the OCLC/Geac 
 Link, which was developed by our head catalog librarian working with 
 Geac systems staff, contains the OCLC number and other identifiers such 
 as LCCN and ISSN numbers. In this way we hope to refine output to reflect 
 better new accessions. While on the one hand we would have wished an 
 easier way to get at new accessions, on the other we are thankful for the 
 flexibility of the Geac system to allow local GLUG programming. 
 
 In an academic setting, class reserves are an important component of a 
 circulation department. One of the strong points of the Geac program 
 when we were looking at circulation systems was its reserve system. I want 
 to talk about two reserve-connected reports: the Weeding and Stamping 
 List and the Reserve Use Report. 
 
 The Weeding and Stamping List is not used at the University of 
 Arizona, but it is an interesting report. To our knowledge it was developed 
 at the request of another Geac library which handles class reserves differ- 
 ently from many libraries we are familiar with. When a reserve list is 
 entered into the system, the online operator indicates when the items are to 
 go on reserve". This program identifies those items in the Geac database 
 which are to go on class reserve between specified dates. After the program 
 is run, the items are displayed automatically in the system as being on 
 reserve with the appropriate reduced loan periods and reserve fine rules in 
 place. The list can be printed to allow staff either to pull material to place 
 in a special reserve location or to use as a listing of what is currently on 
 
GEAC 81 
 
 reserve. From the perspective of listing and automatic display of reserve 
 status, the program is a powerful one. 
 
 However, because we place reserve material in reserve book rooms, we 
 are reluctant to use this program for fear that there will not be sufficient 
 human resources to pull identified material from the stacks in a timely 
 enough manner before it shows on reserve. We prefer to pull the material 
 first and then place them on reserve, using other Geac reserve functions. 
 
 The Weeding and Stamping List is sorted by call number, and lists 
 typical bibliographic information such as author, title and item number. 
 Also shown is the course name, active date and expiry date. Similarly the 
 program will automatically remove material from reserve when the expiry 
 date is reached. 
 
 Another component of management information for class reserve is 
 Geac's program that produces the reserve use report. The standard report is 
 sorted by course and then by professor's name. As one can see, the course is 
 listed only once with the professor's name directly under. Bibliographic 
 information is listed by material type within professor's name, and then by 
 call number, author and title. This is followed by the item barcode number 
 for each item, the item location (this could be different from item to item if 
 a library did not have a centralized reserve room), material type, circulation 
 count, and reshelve count. Through local programming, other sorts are 
 possible such as a call number within a course. 
 
 As designed, the report is useful in giving professors information on 
 what items placed on class reserve are used, and in many cases not used at 
 all. Most of us know that the information will have little impact in putting 
 together next year's class reserve list. We view this report as we do filling 
 out income tax forms. The process probably would not go far in helping 
 us, but we are surely damned if it is not filled out and mailed on time. 
 
 A major problem with the concept of this report is that one cannot 
 delete material from reserve before running it. In doing so, the circulation 
 counts for the reserve status are lost. It is not that the count disappears, but 
 simply that since (presumably) the item is back in the regular stack loca- 
 tion, the circulation information will only show as general circulation. 
 Another real problem with the report, as we understand the system, is that 
 circulation counts stop when 256 circulations are reached. We know that 
 many photocopied items circulate more than this amount. Thus, circula- 
 tion counts for reserve courses are not necessarily accurate. 
 
 Before moving into the financial areas of the Geac system, I want to 
 talk a bit about the TSTA Report I mentioned very early in this paper. At 
 first glance one might think that this reports total circulation counts and 
 other statistics by hour of day. But, as I said earlier, it is not so! The value of 
 this report is not to be found in its ability to count circulation, because 
 renewals are not counted by any standard Geac program. 
 
82 WAYNE MULLIN 
 
 What is valuable is that one can tell circulation (minus renewals), 
 discharges, RRPs, fine activity, and holds placed by hour of the day for 
 each location defined within the system. Currently at the University of 
 Arizona, statistics are shown for the Science-Engineering Library, the 
 Main Library, Government Documents Department, the Library Science 
 Library, the music collection, the Arizona Health Sciences Center Library 
 (which is sharing the system with us) and for the two major reserve book 
 room operations in the main and science-engineering libraries. The report 
 totals each activity by hour and by the day, as well as cross tabulating each 
 activity by location. All totals cumulate monthly when a new count begins. 
 If this report counted renewal activity, it would give a complete picture of 
 major circulation activity for each hour of every day the library was open. 
 Currently, the value of this report is in demonstrating first-time circula- 
 tion and other activities that are useful in planning desk coverage (for 
 example, the assumption that renewal activity parallels circulation activ- 
 ity). A program is being devised to add all of these figures and graph the 
 results by hour. 
 
 A separate area of this report gives the same activity breakdowns by 
 patron statistical class. Remember, statistical class can define a narrow 
 category of users. For example, by looking at the report for 14 April 1982, 
 one can see that UAFRESH (University of Arizona freshmen) checked out 
 7675 items system-wide, returned 6205 items system-wide, had 145 fines 
 issued and placed 128 hold requests for a total month-to-date activity of 
 14,155. The latter figure is a mixture of "apples and oranges," but it does 
 give one a figure to compare with other statistical classes to demonstrate 
 library use or at least circulation-related use. Except for not counting 
 renewals a gross omission it is a very good report. 
 
 Before entering the turnkey circulation marketplace, Geac was heavily 
 involved in banking systems, and still is. I say this because we are now 
 entering into that portion of Geac's circulation system which deals with 
 fines, bills for replacement and the like. Very detailed information is 
 provided online for each user having a fine or replacement bill, including 
 the ability to manually or automatically block usage of the system by a 
 patron when financial obligations reach certain levels. Although strong in 
 the main, the program does have quirks at least as perceived by the staff at 
 Arizona. 
 
 First, one has the ability to modify a billed amount downward, but 
 there is no corresponding ability to modify it upward. As a matter of 
 policy, all bills for replacement are issued at $35 per item. However, if a 
 patron challenges this amount, the fines office staff will seek a price for the 
 item with the understanding that if it is more that $35, the higher amount 
 will be billed. To do so, and it is done very infrequently, is difficult. Either 
 one has to discharge the item in question then manually input "fake" 
 
GEAC 83 
 
 charge-out-and-return information and then input the price, or one can 
 add on the higher amount in the processing fee field. 
 
 Second, in calculating the amount owed to stop (block) users from the 
 system, Geac as standard policy counts only fines not bills for replace- 
 ment. I mention this to give you some idea of areas in which online turnkey 
 reality differs from local reality. The financial programs, though, both 
 operationally and report-wise, are essentially good. 
 
 One of the benefits the University of Arizona was looking for from an 
 online system was an interface between the system billing a user for 
 replacement of a long overdue item and somehow notifying the fines office 
 staff that the particular item had been returned if payment had already 
 been received for the item. Such an interface is available in the report titled 
 LBPATE. This program produces a list of all patrons who are eligible for 
 refunds. It is a report we have not often used, but will do so in the future. 
 The report can be sorted by item barcode number and by patron name. 
 Both are valuable. The information contained in the program includes 
 patron ID number, patron name, return date, item barcode number, and 
 amount of refund. 
 
 One may think that such a report as LBPATE is not necessary. I say 
 this because that was told to me by certain loan staff. The theme went 
 something like this: No one who has paid for a book and then returns it 
 will let the matter drop without coming in and demanding (not asking, 
 mind you) for a refund. Not so! I am saying that some of our assumptions 
 about how users act or react are not necessarily true. There are certain kinds 
 of users who do not come into the library and inquire about refunds. 
 However, the LBPATE report also serves another purpose. It provides the 
 library the opportunity to make a refund when a book paid for by a patron 
 simply shows up. While I do not like to admit it especially for 
 publication I guess we do make mistakes from time to time, and both 
 staff and patron do not discharge material correctly or manage to shelve it 
 in the wrong place. This report gives us an opportunity to refund a fine 
 when the error has been ours. We like this report very much! 
 
 When a library collects money and/or modifies fines, financial man- 
 agement information is of paramount importance. The Geac system does 
 some interesting things in this area, not all of which work in a library 
 setting such as at the University of Arizona. There is excellent online 
 information by patron query and by bibliographic query on fines, missing 
 status, and billing for replacement items. For example, if I were billed for 
 the late return of the book Old Man and the Sea, the information would 
 show on the Bibliography Query and also on my patron record. However, 
 the minute I pay the fine things become a bit confusing. I say this because 
 staff in the library fines office often have to refer back to the Geac-produced 
 Fines Journal to document payment or other action. The journal is pro- 
 
84 WAYNE MULLIN 
 
 duced as part of overnight processing. It lists all activity of the financial 
 module for the previous day and shows, for example, payments, modifica- 
 tions, cancellations, and refunds. The problem and it is a big one is that 
 information is sorted (printed) by time of day. Thus, if I paid a bill two 
 weeks ago, but didn't quite know when or what day, the staff might have to 
 look through all Fines Journals scanning each day for my name. 
 
 To overcome nervous breakdowns by fines office staff and to protect 
 me from attack, our head computer operator asked Geac to report the 
 information by patron name. One still has to look through daily journals, 
 but it is much easier with a name sort than the time of day sort provided by 
 the standard report program. 
 
 One of the nonstandard reports that we occasionally use is the print- 
 ing of financial records for a single user. Again, we have written a GLUG 
 program which will print all financial obligations for a library patron. 
 This is useful when dealing with users who have a lengthy record and who 
 want a hard copy with which to work. The report information is in no 
 particular order, but it does include call number, title, author, item 
 number, charge out, and due date information as well as returned date and 
 time as applicable, and fine amount for each item returned late. The 
 GLUG program also prints out a list of what material a person has checked 
 out. We do not advertise these reports, but rather refer users to a public 
 query terminal to look up their own records. I want to emphasize again 
 that there are many reports that can be "programmed" locally without the 
 services of a formally trained computer operator. I will discuss this in 
 greater detail in the section on Bibliographic and Patron Extract modules. 
 
 One of the things we did not do well (in fact we didn't do it at all) was 
 to specify in our RFP that the Geac system have a program to transfer 
 financial information from the system to the university business office. I 
 mention this to point out that one cannot assume a vendor will provide 
 what may seem obvious. 
 
 The University of Arizona has a policy of collecting fines in the 
 library. However, according to state law all fine amounts must be trans- 
 ferred to the business office so that student records can be encumbered. 
 Geac does have a program to do this, but in our opinion it is so tailored to a 
 specific Geac site that it is of no use to us at all. The Geac-developed 
 program dumps aggregate financial information onto the tape for transfer 
 to a business office. In doing so, however, all library financial online 
 records are wiped clean. This was unacceptable to the University of Arizo- 
 na Library because we are more interested in getting material back through 
 talking with users than we are in collecting money. We are now working 
 with the Geac system and the university's business office to develop a taped 
 snapshot of student financial records to transfer into the university's 
 encumbrance system while still retaining financial information in the 
 
GEAC 
 
 Geac database. It is important to note that this capability is above the cost 
 of the initial system. 
 
 The Geac does provide the capability to print out total financial 
 records. In fact the library was asked to run this report in lieu of having a 
 tape transfer program in place. We did this with humorous results at 
 least from our point of view. After an hour or so, we were able to begin 
 printing. When we were 1 percent of the way through printing, we had a 
 40-page output and the operator made a few quick calculations. Determin- 
 ing that he would be there ten hours, he terminated the program. Clearly 
 the report was possible to produce, but our estimates were that the business 
 office would have a pile of paper over three feet high. Upon notification of 
 this reality, they decided to overlook the fact that student library fines 
 would once again not be part of the encumbrance system. 
 
 This report, like some others that Geac provides, can be run more 
 reasonably in a small institution with small patron (bibliographic) data- 
 bases. It would certainly be difficult to run such a report in a library the size 
 of the University of Arizona's. 
 
 The report, I think, would work very well with the standard transfer 
 program Geac has developed as part of the turnkey system. A library could 
 run the patron financial report prior to running the transfer program. In 
 doing so, staff would have a record of patron financial information, 
 although it would mean going through each report to locate a specific 
 patron record. 
 
 The last report I want to discuss in the financial area is Geac's 
 standard Long Overdue Summary Report. It lists items overdue beyond a 
 time period defined by the library. The report is sorted by call number and 
 displays the patron name, item number, call number/author/title, due 
 date, return date, fine amount, price, and type. The report has as its main 
 use, we think, that of searching for overdue material. It is another report 
 that is of little utility to the University of Arizona for a number of reasons 
 which may not apply to other libraries. Once again our size and policy/ 
 procedures are against us. The report is cumulative and reports all items 
 fitting the time frame profile. The potential problem with this report, as 
 we understand it, is that over time it would become too large for a limited 
 staff to use for searching or even notifying users in an attempt to retrieve 
 material. Currently at the University of Arizona, there are over 1200 items 
 represented on this report. 
 
 I would like now to discuss that portion of the Geac system with which 
 I personally have the least familiarity because the programs are executed by 
 computer operators only Bibliographic and Patron Extracts. The Biblio- 
 graphic Extract is a statistical tool which allows one to design reports for 
 most bibliographic entries in the system. When looking at it, it is probably 
 more important to explain what cannot be done rather than what can be 
 
86 WAYNE MULLIN 
 
 accomplished. Outstanding limitations of the system include: (1) biblio- 
 graphic entries that do not have copy level information cannot be accessed; 
 (2) current transaction data is not accessible, and historical transaction 
 data is limited; and (3) the circulation counter does not count renewals. 
 The operator can define rules using fields either at the copy (item) level of a 
 record or at the bibliographic level (data relevant to all copies of the title). 
 One may decide to either count the number of matches for the rules defined 
 or else to print a list of all items that matched the rules. Geac does not sort 
 the data to be printed, but the sort can be easily arranged. 
 
 Although, in theory, the Bibliographic Extract can combine almost 
 any number of rules in any combination of item and title level informa- 
 tion, we have found that, in reality, certain requests cause the program to 
 run so slowly that it is greatly reduced in effectiveness. As an example, it 
 runs particularly slow when asked to look for certain call number ranges 
 e.g., all call numbers beginning with QA. Although we had hoped to use 
 the extract program to count collection growth, such programs take too 
 long to run. We can ask for the number of books in a particular agency 
 which have circulated more than N times, but we cannot ask for books that 
 have circulated to a specific patron class more than N times, because this 
 information is not stored in files the extract can access. We can also ask for a 
 list of books in a specific location which have not circulated since a 
 specified date, but may not ask for a list of books currently charged out. 
 Despite its limitations, we find the Bibliographic Extract to be extremely 
 useful. Data which may be defined includes, but is not limited to, the 
 following: 
 
 At the bibliographic level, deleted titles, titles which are part of a 
 multi-volume set, data-verified titles, reordered titles, titles added online, 
 titles with notes, titles bound with other titles, number of pending holds on 
 a title, call number/author/title of a work, notes added to a bibliographic 
 record, publisher of a title, subject of a title, system accession number, date 
 the record was created, date of last update, ISBN number, LCcard number, 
 library-defined optional fields (8), OCLC number, local accession number, 
 the price of a book, and the number of copies. 
 
 At the item level, the data which may be defined includes the material 
 type, whether an item is missing and the missing type, the item's normal 
 location, whether a copy has been deleted, whether a copy belongs to the 
 library, whether a book is missing, whether an item is on reserve and 
 reserve type, copy is bound with, whether it is fine type, the number of 
 times the copy was charged out, number of times copy was discharged 
 when not previously charged, copy barcode number, transaction indicator, 
 system reference number, date of last change, local copy specific call 
 number, local copy number, secondary location, and library defined 
 optional fields (4). 
 
GEAC 87 
 
 The Patron Extract works much like the Bibliographic Extract, with 
 many of the same virtues and limitations. One may define a number of 
 rules, then either list or count the number of patrons who match those 
 rules. The Patron Extract gives more access to the transaction file, but does 
 not allow one to print any specific items charged to any specific patron. 
 Because reserve courses are actually pseudo-borrowers, the Patron Extract 
 also allows some access to the reserve subsystem. Information which may 
 be defined includes, but is not limited to, the following: 
 
 the borrower's ID number; 
 
 whether the borrower's privileges have been suspended and a notice sent; 
 
 whether the borrower has recalled books and a notice has been sent; 
 
 whether the borrower has been sent an overdue notice; 
 
 if the borrower has overdue books; 
 
 whether the borrower has an informative message in his record; 
 
 if the borrower has an incomplete address (determined by the library); 
 
 if the borrower has satisfied lapsed or terminated holds; 
 
 whether a borrower's badge has been reported lost, missing or stolen. 
 
 An expanded list is included as an appendix to this paper. The last area of 
 management that will be discussed are the Geac system's management 
 capabilities. 
 
 Geac takes care of itself very well; in some ways better than it does 
 those of us who seek more from it than it can supply. Geac provides 
 excellent system management information which is generated in the over- 
 night processing routines. These routines are done most nights at the 
 University of Arizona. Very briefly I want to discuss some of the informa- 
 tion contained in these reports, all of which to me have very strange names. 
 
 One of the most carefully watched reports is LPFCHK which, among 
 other things, tells us the growth of various files in the system. I frequently 
 watch the percent column of the report which states how full each file is. 
 Major files contained in this report include text, history, item, transaction, 
 patron, and interagency. The computer operators not only are concerned 
 with how full files are, but the rate of a file's growth. We pay particular 
 attention to the history file because it tends to overflow rather easily. 
 
 The TSTA report has been discussed previously. I mention it here 
 only to point out again that it is generated in overnight processing and 
 could be considered a systems management report. The Terminal Line 
 Analysis Report is one that only a computer nut or a very statistically 
 minded librarian would enjoy. It is Geac's way of telling us how hard it has 
 worked any given day. The important information contained in this report 
 falls into two areas: port activity and terminal activity. Each port in the 
 system is listed with the number of sends (each time someone hits the send 
 key and a command goes to the computer) and the total number of 
 
88 WAYNE MULLIN 
 
 characters (the nonhuman kind) represented by the sends, which are also 
 known as GODO counts. The first portion of this report totals the number 
 of sends and the number of characters sent. Looking further, one can find 
 out how active each specific terminal was for the day in question by 
 looking at the terminal detail portion of the report. Contained here is the 
 terminal number, the number of sends for the terminal, the number of 
 characters represented, the number of wands represented (my favorite 
 statistic), the number of wand errors (most always zero), and the number of 
 times the terminal had to be reset. I exaggerate somewhat when I make fun 
 of this information, for what one has here is the ability to monitor terminal 
 use. For example, by knowing where terminal number ten is located, one is 
 able to see (perhaps) that it is not getting much use and should be moved 
 elsewhere. The decision is always being made, of course, by someone other 
 than I. As a matter of possible interest, we find that public query terminals 
 get the most use. 
 
 The reset statistics prove valuable in two related instances: when the 
 University of Arizona first came online and also as we brought new sites 
 onto the system. One most often has to reset (cancel out a screen display and 
 bring back the general menu) when one does not know what to do next. By 
 watching reset activity, loan department training staff were able to zero in 
 on potential problem sites. Of course we did not know what the problem 
 was, but we could pretty well be sure one was present. In addition, this 
 report is essential to the recovery process should the system have a fatal 
 shutdown. 
 
 LPOVR is a rather cumbersome report which, among other things, 
 gives the total borrower records, item records, and call number records in 
 the database. This report also displays summary details of notices sent. 
 However, this is not to be confused with individual pieces of paper pro- 
 cessed (in our case, stuffed into envelopes), but rather totals for the different 
 kinds of notices generated. We, like many libraries with Geac systems, use a 
 multipurpose form on which a variety of different kinds of notices can be 
 found e.g., on the same piece of paper can appear an overdue fine, a recall 
 and first overdue date. 
 
 Areas covered statistically for notices generated include totals for 
 number of records output (the total of various kinds of notices); bills for 
 replacement created; processing fees created; the number of first, second 
 and third overdues; fines; recalls placed including reserve recalls; available 
 notices; cancelled holds; expired holds; number of term notices sent (at 
 Arizona this represents faculty and staff reminder notices); processing fees; 
 replacement bills; number of items sent for collection; and finally, the 
 number of long-overdues generated. Other useful information provided by 
 the LPOVR Report is the total dollar amount of overdue fines payable, the 
 total of fines owed (overdue but not yet returned, thus the fine is not yet 
 
GEAC 89 
 
 totaled), and the total overdue fines and fines owed. By using this report, 
 one is able to get a general idea of notice and financial activity for the day in 
 question. Early on, I spent each morning going through all the overnight 
 processing reports. 
 
 One report that is very useful, particularly with regard to having the 
 Geac comply with response time mandates in our contract, is a report that 
 lists terminal response times. It is a complicated report to understand, but 
 Tom Owens tells me Geac personnel have been very patient in their 
 explanations. We look at this report daily for signs of response problems. 
 
 The final Geac looks-at-itself report I want to present is one that gives 
 what I call demographic statistical information. Displayed are the number 
 of transactions and borrower records on file as well as transaction counts 
 for the following: charges, discharges, fines (lumped together); holds; 
 RRPs; financial; partial payments; current books on loan; current books 
 on term loan; current books on reserve; and missing books on loan. Other 
 information presented may be of interest to a computer operator, but has 
 little value to me. Included are such things as data on the longest transac- 
 tion chain, number of financial chains, and so on. 
 
 We have now completed looking at the four areas of management 
 information one can obtain from the Geac system: circulation, financial, 
 bibliographic and patron extract, and how the system manages itself. If I 
 had time this afternoon to make two points only, they would be that there 
 may not be such an animal as turnkey reality, just different local realities 
 that have needs which may or may not be met with a system called 
 "turnkey"; and always ask for the obvious and then always expect the 
 unexpected when dealing with the turnkey marketplace. 
 
90 WAYNE MULLIN 
 
 APPENDIX 
 
 Data Elements for the Bibliographic 
 and Patron Extract Module Programs 
 
 In Bibliographic Extract Module programs, data which may be defined include the 
 following categories: 
 
 At the Bibliographic Level: 
 deleted titles 
 
 titles part of a multi-volume set 
 data verified titles 
 reordered titles 
 titles added online 
 titles with notes 
 
 number of pending holds on a title 
 call number, author and title of an item 
 notes added to a bibliographic record 
 publisher of a title (when data available) 
 subject of a title (when data available) 
 system accession number 
 date record was created 
 date of last update 
 ISBN number 
 LC card number 
 
 library defined optional fields (eight available) 
 OCLC number 
 local accession number 
 price of book 
 number of copies 
 
 At the Item Level: 
 material type 
 
 item is missing and missing type 
 item in normal location 
 copy has been deleted (CPY*DEL) 
 book is missing 
 
 item is on reserve and reserve type 
 copy is a bound with 
 copy has fine type 
 number of times copy was charged 
 
 number of times copy was discharged when not previously charged 
 copy barcode number 
 transaction indicator 
 system reference number 
 date of last charge 
 local (copy specific) call number 
 local copy number 
 secondary location 
 library defined optional fields (four available) 
 
 Boolean strategies within these categories can also consider the reverse meanings as 
 well (e.g., listing all items with three or more pending holds but without a note 
 added to the bibliographic record, if so desired). 
 
GEAC 91 
 
 In Patron Extract Module programs, similar boolean strategies can be pro- 
 grammed. Information which can be accessed by the Patron Extract includes (but is 
 not limited to): 
 
 borrower's ID number 
 
 borrower's privileges have been suspended and a notice sent 
 
 borrower has recalled books and a notice has been sent 
 
 borrower has overdue books 
 
 borrower has been sent an overdue notice 
 
 borrower has an informative message in his record 
 
 borrower has an incomplete address (as determined by the library) 
 
 borrower has satisfied, lapsed or terminated holds 
 
 borrower's badge has been reported lost, missing or stolen 
 
 borrower has an overdue recalled book 
 
 borrower's privileges have been stopped because a book is too long overdue 
 
 borrower's fines are over the limit set by the library 
 
 borrower's privilege class 
 
 borrower's record was created online 
 
 borrower has transactions 
 
 borrower surname 
 
 borrower initials 
 
 borrower note text 
 
 number of books a borrower has on loan 
 
 number of books a borrower has been billed for 
 
 number of reserve loans for a borrower 
 
 both borrower's home and local address 
 
 borrower's phone number 
 
 borrower's statistical class 
 
 date borrower was registered 
 
 date borrower's current validation expires 
 
 borrower's barcode number 
 
 library defined optional fields (six available) 
 
 borrower's agency 
 
 reserve borrower's course name 
 
 reserve borrower's professor's name 
 
 number of borrower's current and/or deleted transactions 
 
 number of CDFs (charges, discharges, fine transactions) for the borrower 
 
 number of items on reserve for reserve patrons 
 
 number of financial transactions for the borrower 
 
 number of hold transactions for the borrower 
 
 number of partial payments of fines the borrower currently has 
 
 number of unpaid fines the borrower has 
 
 number of fines the borrower has which have relevant correspondence 
 
 number of inactive fines a borrower has 
 
 number of fines available for refund for a borrower 
 
 number of overdue fines for the borrower 
 
 number of recalls for the borrower 
 
 number of processing fees accessed by the library 
 
 number of refunds the library has given the patron (For example, the Patron 
 
 Extract system can tell us how many patrons have more than twenty unpaid 
 
 fines, but not how many patrons owe more than $30.) 
 
JOHN N. OLSGAARD 
 
 Graduate Student 
 
 Graduate School of Library and Information Science 
 University of Illinois at Urbana-Champaign 
 
 Characteristics of Managerial Resistance 
 to Library Management Information Systems 
 
 Most of the presentations given at this clinic have made two assumptions: 
 that library managers understand the functioning and capabilities of 
 automated systems; and, given that they understand the system, that they 
 will utilize the information generated by these systems. These may not 
 always be valid assumptions. To illustrate this fact, several years ago I was 
 asked to be a member of a library systems study team. As part of my duties 
 during the course of that study, I analyzed the patterns of questions that 
 had been asked hourly at the main reference desk for the previous years. 
 After considerably torturing the data, I developed a prediction model that 
 was intended to serve as a guide for adequately staffing the reference desk at 
 any particular time. The lengthy report I submitted contained graphs, 
 tables and explanations of the entire process. The pride of this report was a 
 day-by-day weighted listing of reference activity gauged by the time of the 
 academic year. About three weeks after I submitted this portion of the study 
 to the library director, he asked me to come back for a conference. As I 
 walked into his office, he had my three foot long day-by-day graph layed 
 out on the table, studying it intently. The director was repeatedly saying, 
 "Beautiful, just beautiful." Then he looked up at me, smiled, and said, 
 "But what does it mean?" After attempting to explain the significance of 
 the report, I had the feeling that he still thought it looked pretty, but had no 
 real meaning. 
 
 The point of this story is that management information systems are 
 worthless if the manager either does not understand the information, or 
 refuses to use the information. This is particularly relevant since for the 
 past decade libraries have shown a determined interest in the prospects of 
 automating various clerical processes. Only recently has this interest 
 
 92 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 93 
 
 begun to include management processes as being applicable to library 
 automation. In a recent issue of College & Research Libraries, Robert S. 
 Runyon suggested that library management is now ripe for automated 
 administrative information gathering systems. 1 
 
 Runyon is no doubt correct in his belief that the library management 
 profession will be examining the concept of management information 
 systems (MIS) more closely in the future. However, his thinking is based on 
 the assumption that library managers will use the systems once they are 
 available. This paper is founded on the premise that in most libraries the 
 current information that is derived from automated systems is a nonuti- 
 lized or underutilized commodity in the management process. The pur- 
 pose of this study is to examine the characteristics of this failure to use 
 management information. This examination will consist of three major 
 sections: 
 
 1. An examination of how the library as an organization has been 
 responding to a changing environment, both externally and internally. 
 
 2. An investigation of the causes of the nonuse of information derived 
 from automated systems in the decision-making process. 
 
 3. An analysis of the rationale for using this kind of information in the 
 management process. 
 
 Organizational Reaction to External and Internal Stimuli 
 
 Most libraries have a problem keeping their clients totally satisfied 
 with the service their organization provides. One almost comes to expect 
 the regular grilling in the newspapers over what the libraries are, or are 
 not, doing to our unsuspecting public. Further, the litany of internal 
 employee problems has become as common as news of another budget cut. 
 The library in today's environment faces what those in management circles 
 might call a marketing problem. 
 
 Most marketing problems are really exchange problems. A diagram of 
 the exchange process is given in figure 1 . For example, if you go to eat in a 
 university cafeteria the obvious exchange would be the cost of less person- 
 alized service and uncomfortable seating for the benefits of fast service and 
 no tipping. 
 
 External Stimuli 
 
 On the external exchange level, organizations can be seen, in a natu- 
 ralistic sense, as having a symbiotic relationship with the community they 
 serve. Each element the organization and the community have a mutu- 
 ally beneficial effect. The organization provides a service needed by the 
 community, and in return the community supports the organization. The 
 
JOHN N. OLSGAARD 
 
 ORGANIZATION 
 
 COSTS 
 BUNDLE 
 
 BENEFITS 
 BUNDLE 
 
 CONSUMER 
 COMMUNITY 
 
 Fig. 1. Representation of the Exchange Process in Marketing 
 
 problem is that this relationship is not static. As the environment changes, 
 both the community and organizational needs change. When the service 
 needs of the community either out pace the ability of the organization, or 
 when the service needs of the community cannot be fulfilled by the organi- 
 zation, we perceive what Miriam A. Drake has termed performance gaps. 2 
 The first symptom of a performance gap is an organizational version 
 of stress. Miller has suggested that stress occurs in an individual when 
 either there is a lack of some essential input, or when an excess of input 
 floods the system. 3 Meier has taken this concept a step further by suggest- 
 ing that the same kind of stress experienced by individuals can affect 
 organizations. Meier postulates that as the imbalances that occur between 
 the demand for an organization's services and that organization's ability to 
 deliver those services increases, the amount of stress on the organization 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 
 
 will increase. 4 As the imbalance grows more severe, the organization 
 experiences the equivalent of a nervous breakdown. The organization 
 simply stops functioning. This phenomenon is illustrated in figure 2. 
 
 DEMAND 
 
 Fig. 2. Example of the Affects of Stress on an Organization as Demand Increases 
 (Modified from Meier, Richard L. "Information Input Overload: Features of Growth in 
 Communications-Oriented Institutions. In Mathematical Explorations in Behavioral 
 Science, edited by Fred Massarik and Philburn Ratoosh, p. 268. Homewood, 111.: Dorsey 
 Press, 1965.) 
 
96 JOHN N. OLSGAARD 
 
 For example, those of you who are familiar with academic libraries 
 may recognize this kind of deterioration occurring during the weeks 
 preceding final examinations. As the demand for services of the academic 
 library increases, it is met with increased performance from the library's 
 staff. However, the situation reaches the point where the staff is perform- 
 ing at its maximum level while the demand for service continues to 
 increase. This is typified by overcrowded study areas and long lines at the 
 circulation and reference desk. The staff quickly discovers that no matter 
 how hard or how fast they work, their efforts will not keep up with 
 demand. It is at this point that the organization gives up trying to keep 
 pace with demand; the staff either sets its own performance pace regardless 
 of the demand, or begins a policy of high absenteeism in an attempt to 
 ignore the demand. 
 
 It is the library manager's job to ensure that this kind of breakdown 
 does not happen in a library organization. When a performance gap begins 
 to occur between the library and its community, the library manager must 
 either seek a new community or alter the library to provide new services. 
 The essential problems that the library manager must solve are: (1) to 
 recognize that an imbalance or performance gap exists, and (2) to know 
 what direction the organization should move to correct the imbalance. It 
 can be suggested that the past performance record of library managers in 
 solving these kinds of problems has been something less than totally 
 successful. 
 
 Roger Horn has suggested that this poor performance record for 
 libraries is attributable to what he believes to be the generally poor quality 
 of administrators libraries attract. 5 Without disclaiming that there are 
 poor library managers in the field, or that they have committed some truly 
 magnificent decisional blunders, one hopes that there are other reasons for 
 this failure other than lack of talent. 
 
 Internal Stimuli 
 
 Many library managers and automation specialists believe that since 
 they do not have much contact with the general public, this exchange 
 process really does not apply. However, in a similar way, the exchange 
 process occurs every day within the organization. For automation special- 
 ists, the primary consumers are probably library managers; for library 
 managers the direct consumers of their product (i.e., administrative deci- 
 sions) are other librarians within their organization. The basis of the 
 exchange for library managers are fair administrative decisions in return 
 for organizational power and, presumably, loyalty from their employees. 
 These intraorganizational exchanges are as important as the exchange 
 between the organization and the ultimate user community. It could 
 further be submitted that there has been a history of unnecessarily limited 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 97 
 
 cooperation in these intraorganizational exchanges. Automation special- 
 ists and library managers, as respective groups, have based their products 
 on an internally-oriented viewpoint. That is, these respective groups have 
 been so interested in the manner, the form and the process of how their 
 products are generated that they tend to forget how the products are 
 utilized. 
 
 Alan R. Andreasen has devised a checklist of "yes or no" questions to 
 determine if an individual's or institution's service is too internally 
 oriented: 
 
 1. Is customer ignorance the barrier to the success of your product? In 
 other words, if people weren't so stupid, they would see the importance 
 of your service. 
 
 2. Is your product inherently good? 
 
 3. Do you view the consumer as "the enemy?" If the people who use your 
 service would just leave you alone everything would be perfect. 
 
 4. Do you see your marketing problem as changing consumer rather than 
 changing the product? 
 
 5. Is communication the only really important marketing tool needed to 
 push your product? 
 
 6. Do you believe consumer research isn't really necessary? 
 
 7. Do you believe consumers of your product are all the same? 6 
 
 If the answer is "yes" to all or most of these questions, the institution's 
 service is too internally oriented, and probably most librarians and librar- 
 ies suffer from this malady. On the one hand the automation specialists are 
 justifiably proud of the kind of timely information services they can 
 provide, but consider themselves hindered by the library managers that use 
 their product because the managers seem incapable of understanding the 
 most basic data computations. On the other hand are the library managers 
 who view the automation specialists as some kind of vague reincarnation 
 of Dr. Strangelove those who relate well to their machines, but have little 
 grasp of the real world. 
 
 What has occurred in the case of library managers and information 
 specialists is a failure on the part of both parties to properly understand the 
 exchange process involved. Similarly, the automation specialist wants the 
 benefits (i.e., a job and good facilities), without attempting to understand 
 the kinds of services or costs that will be required. What others have called a 
 "breakdown in communications" is what Meltzer claims is actually a loss 
 of information. 
 
 This internal information exchange can perhaps be best explained by 
 use of diagrams (see figs. 3 and 4). In figure 3 one can see the familiar 
 pattern of a typical organizational chart. However, a better way to view the 
 communication between the various administrative levels can be seen in 
 
98 
 
 JOHN N. OLSGAARD 
 
 MANAGER 
 
 MIDDLE 
 MANAGER 
 
 MIDDLE 
 MANAGER 
 
 LINE 
 PERSONNEL! 
 
 LINE 
 PERSONNEL 
 
 LINE 
 PERSONNEL! 
 
 LINE 
 PERSONNEL 
 
 CLIENT COMMUNITY 
 
 Fig. 3. Example of a Typical Organization Structure 
 
 figure 4. The manager only has direct communication with the middle 
 managers, the middle managers directly communicate with line person- 
 nel, and line personnel with the clients. The manager has knowledge of 
 what clients are interested in or what problems they are experiencing only 
 as that information is filtered through the organization's line personnel, 
 and refiltered through middle management. Concomitantly, an adminis- 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 99 
 
 CLIENT COMMUNITY 
 
 Fig. 4. Representation of Communication Between Organization Levels 
 
 trative decision to change the direction of the organization goes through 
 somewhat the same filtering process as it progresses outward from the 
 manager; the decision at each level is filtered, interpreted and readjusted. 
 This situation is rather common; at each level the basic information is 
 reformulated in terms of the individual desires and political considerations 
 at that level. At each additional level the person who needs to make the 
 decision only has the word of the people at the earlier level that the 
 situation they are describing is accurate. For instance, the people at the 
 manager or middle manager level have become desensitized to increased 
 
100 JOHN N. OLSGAARD 
 
 budget requests because that tends to be the only kind of information they 
 receive. Managers have found that there are very few instances where 
 employees claim that they have too much of anything or wish to have 
 something cut from their budget. 
 
 The basic problem with this loss of administrative information 
 between the various organizational levels is that the decision-makers in the 
 library have no reliable benchmark with which to gauge the importance of 
 the request. As such, libraries rarely make informed organizational deci- 
 sions to meet the needs of their clients or employees. The library must be 
 able to legitimately view itself as a dynamic entity that can successfully 
 react to its changing environment. 
 
 The internal exchange failure within a library can exacerbate an 
 external exchange imbalance. For instance, if the public services staff 
 reports to the library manager that demand for a certain subject is outstrip- 
 ping the collection, the library manager must decide if the information is 
 accurate enough to warrant a change of policy. This is where information 
 derived from automated systems could play a major role. Most automated 
 circulation systems can generate a frequency list of circulated materials. In 
 addition, many automated acquisition systems can produce similar fre- 
 quency lists. Thereby, the library manager can use the circulation informa- 
 tion as a benchmark of the demand for a certain subject, and the 
 acquisition information as a gauge of how well the library is responding to 
 the demand. The library manager can then make a rational decision on 
 whether the demand requires an additional appropriation. 
 
 In many libraries these automated systems already exist, and in most 
 cases a byproduct of the systems include the above mentioned report 
 features. Unfortunately, library managers for the most part have refused to 
 incorporate these channels into their decision-making plan. Understand- 
 ing the causes of nonuse of information derived from automated systems 
 constitutes a major dilemma for library organizations. 
 
 Causes of Nonuse of Information Derived from 
 Automated Systems in the Decision-Making Process 
 
 The general causes of managerial resistance to information derived 
 from automated systems can be loosely grouped into four categories: 
 (1) educational deficiencies, (2) sensory and decisional deficiencies, 
 (3) cultural deficiencies, and (4) the "priesthood effect." 
 
 Educational Deficiencies 
 
 Much of the blame for the failure of library managers to utilize 
 automated data has been attributed to the inability of managers to under- 
 stand the largely numerical-based format of the data, and admittedly, the 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 101 
 
 inability of software manufacturers to create formats that make this kind of 
 data easier to understand. Shera has suggested that a sixth year of study be 
 added to library science programs to make up for these educational defi- 
 ciencies. 8 Divilbiss goes one step further to recommend the recruitment of 
 undergraduate science majors into the profession. He points out that many 
 librarians not only have little previous academic training in fields that 
 lend themselves to fully understand automated systems, but generally 
 receive little help in this area when they reach library school. 9 This gener- 
 alization can perhaps be expected. After all, librarians are either directly 
 involved with, or are the product of, the educational process. As such we 
 tend to view solutions to many problems in an educational light. 
 
 Although the educational deficiencies of librarians certainly have a 
 part to play in the nonuse of data derived from automated systems, it is 
 probably too glib to rest the entire explanation on this factor. A strictly 
 educational causation would lend itself to a relatively simple solution 
 get the necessary education through a course or two in statistics and 
 computer science. 
 
 One of the foundations of this paper is to suggest that several other 
 factors may be operant in any given situation, but not that all are occurring 
 at the same time or in equal amounts. 
 
 Sensory and Decisional Deficiencies 
 
 The first of these factors has to do with information overload. The 
 automated systems that library managers are or will be dealing with 
 have reached a level of sophistication that they can generate mountains of 
 data at the touch of a button. The problem comes from trying to interpret 
 all of this information in a rational, logical manner. This may be a classic 
 case of what Toffler referred to in Future Shock. 1 There are many exam- 
 ples of where individuals who are faced with increasingly difficult deci- 
 sional tasks will give up attempting to cope with the new stimuli. They 
 simply quit trying to process the new information. The ability of auto- 
 mated systems to generate data has far outreached the library manager's 
 ability to interpret that information and react to it in a useful way. 
 
 David Firnberg has pointed out that when faced with a frustration, 
 like information overload, we react as any animal would when confronted 
 with an obstacle: "the animal either lies down pretending not to notice and 
 goes to sleep; or it rejects the situation, turns its back and walks away; or it 
 battles and tries to master and overcome the cause of its frustration. In our 
 reactions most of us fall into the first two categories." 11 Unfortunately it is 
 rather easy for information overload to occur. Both Posner and Miller have 
 demonstrated the rather severe cognitive limitations of the human mind. 12 
 
 It could be argued that this is merely an extension of the argument that 
 the solution rests in better educational training (i.e., the better theeduca- 
 
102 JOHN N. OLSGAARD 
 
 tion, the better the individual will cope with the information). Melnyk has 
 suggested that those who have been instructed in the use of a computer 
 facility will experience less frustration than those with no training. 13 This 
 would no doubt help. However, it can be suggested that the capacity of 
 such systems would quickly outreach even the highly trained individual. 
 
 Cultural Deficiencies 
 
 One of the less talked about reasons for managers not using data 
 derived from automated systems are various inherited cultural biases. One 
 of these biases would include a basic resistance to machines of any kind, 
 particularly machines whose functioning is difficult to understand. One 
 remembers that during the industrial revolution in England some of the 
 workers, the Luddites, destroyed looms. More recent examples would 
 include the numerous instances of people punching extra holes in com- 
 puter billing cards, or the individuals who input obscene entries into a 
 national cataloging database. In fact, one of the major papers given at the 
 1981 ACRL National Conference was presented by Paul Lacey and was 
 entitled "Views of a Luddite." It would probably be safe to say that the 
 computer inspires as much distrust as any other technological innovation 
 of our day. 
 
 This kind of resistance as demonstrated by managers has been the 
 subject of a number of studies. Ennis has explored the resistance of librar- 
 ians to automation. 15 Others have attempted to identify the characteristics 
 of those managers that are prone to resist automation efforts. A recent 
 Business Week reported the results of a survey conducted by Booz, Allen 
 and Hamilton, Inc. on this topic. 1 It was found that older managers who 
 had been with the same company for a long time tended to be far more 
 resistant than younger executives who had moved from one company to 
 another. 
 
 It is interesting to note that this is the same result postulated by Rose 
 in 1969. Rose theorized that older managers who had a long tenure with the 
 same company tended to be more resistant because their strength and 
 organizational power stemmed from a superior knowledge of the current 
 system. Since automation would disrupt that system, it posed a threat to 
 the older executive. 1 
 
 De Greene has suggested that resistance to automation is not unique 
 to the personal characteristics of managers, but it is unique how automa- 
 tion affects their particular positions within the organization. 18 Managers 
 who were favorably disposed toward automation when it meant the elimi- 
 nation of clerical employees, are now suddenly very resistant when 
 managerial-level positions may be eliminated. 
 
 Aside from the the fear factor of automation, even intelligent people 
 have a basic distrust of computer-generated data. They have a feeling that 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 103 
 
 somehow the machine has made a mistake. Thereby any decisions they 
 would make on the basis of that data would be faulty. There are scientists 
 who will run their calculations on the computer, and then cross-check 
 them on their calculators just to be sure. They know that the odds against 
 the computer making an error of that nature is astronomical, but there is a 
 nagging doubt. The phenomenon of distrusting computer output is de- 
 scribed by Sanders. 19 
 
 A second cultural bias has to do with the nature of operating a 
 computer. Up until a few years ago (and still in many places) one had to 
 feed punched cards into a reader by hand in order to make the computer 
 function. Most systems still require the operating of a typewriter-like 
 keyboard. Many managers resist the idea of having a keyboard terminal in 
 their office because they feel it makes them look like a secretary. One must 
 remember that our culture is one that regards gardening for a living as a 
 lower class activity, but gardening for a hobby is sublime. Ergo, many 
 managers believe that any work done with their hands is simply below 
 them on the sociocultural ladder. Fortunately this problem may be solved 
 by voice-activated terminals. 
 
 Perhaps the most difficult cultural bias to overcome is the prevailing 
 attitude among managers that administration is an art. An art that is 
 simply not conducive to automated data. The environment of manage- 
 ment philosophy has been described by Easton as a broad river valley. On 
 one bank of the river are the managers who believe that management 
 decisions should be made on the basis of experience, intelligence and gut 
 instinct. On the other bank reside the managers who base judgments on 
 decision-trees and computer-generated facts. In truth, of course, a good 
 manager uses a combination of both experience and instinct, and perfor- 
 mance probabilities that are suggested via automated means. The problem 
 is basically one of ego in this case. Many managers feel that by using 
 automated data for decisions they somehow devalue their own self-worth. 
 
 The Priesthood of Automation 
 
 When large-scale computers were just coming into the commercial 
 marketplace, the story was told of a company (we'll call the XYZ Corpora- 
 tion) that had purchased one of these mainframes. The individual whom 
 the company put in charge of this facility immediately had a multitiered 
 glass partition installed around the machine, and established a super-clean 
 climate-controlled environment. Within this partition workers wore head 
 caps and booties. Of course later it was discovered that the computer did 
 not require this kind of special care to operate properly. The operator had 
 simply sought to increase the mystery surrounding the installation for the 
 other members of the company. Thereby no one questioned procedures 
 
104 JOHN N. OLSGAARD 
 
 that were obviously beyond their comprehension. This was empire build- 
 ing in the finest tradition. 
 
 The situation in XYZ Corporation described above may not be as 
 farfetched as it initially sounds. White pointed out that the historian 
 Henry Adams observed at the turn of the century that the machine had 
 replaced the church as an institution of worship in American culture. In 
 his time the dynamo was the extent of technological innovation. In our 
 time the archetype of technology is the computer. 21 If Adams is correct, the 
 computer has all the essential elements that good religion contains. It 
 jointly inspires fear, wonder and pride. Fear, because it is incomprehensi- 
 ble and unknowable to the uninitiated layman; wonder, because its opera- 
 tion approaches being magical; and pride, because although one cannot 
 understand how the machine functions, he knows that it was the inspira- 
 tion of others and fashioned with human hands. To extend the analogy, 
 the initiated have a sacred language known only to themselves. One only 
 has to listen to two computer scientists talking over a problem to recognize 
 this fact. Last, the computer itself can be seen as an icon a symbol of the 
 elect. These initiated, these "elect," are whom I refer to as the "priesthood 
 of automation." 
 
 Like the computer operator in the XYZ Corporation, there are many 
 priests of automation who see this special knowledge as a way to gain 
 power in the organization. Not only does this priesthood not educate 
 management on the functioning of the automated system, but also make 
 sure that management does not gain that knowledge. From this kind of 
 activity, the priesthood can never be wrong, for their argument can always 
 be that management simply is not intelligent enough to see the truth. 
 
 It is easy to argue that the nonuse of automated data rests solely on the 
 head of uneducated managers. It is somewhat more frightening to believe 
 that there could be those within the organization who would intentionally 
 make the system difficult to understand. This is not to suggest that every 
 automation specialist in every organization is doing this. However, it is 
 possible that if a manager cannot get an automation specialist to give a 
 straight answer in plain English, it might be that the person in charge of 
 automation does not want the manager to know the answers. 
 
 Several writers have reflected on the problems of dealing with automa- 
 tion specialists in an organization. Donaldson, Stevens and Becket warn 
 that the automation specialist "thinks of himself.. .as a computer expert, 
 and will regard your business problems as tiresome distractions that come 
 between him and his real vocation." 22 Montague has stated that libraries 
 have brought this problem on themselves by not taking charge of techno- 
 logical innovation. 23 In other words, managers have allowed automation 
 decisions to be made by the very people who have a three-piece vested 
 interest in the outcome. 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 105 
 
 A compromise needs to occur. The automation specialist at today's 
 technological level can build systems that produce data that are timely and 
 understandable. The manager must be willing to pursue the kind of 
 training that will ensure that the data that is produced can be utilized. As 
 Orlicky suggests, the manager must become more of a priest of automa- 
 tion, and the automation specialist must become more familiar with the 
 more profane day-to-day decisions that are made within the organization. 24 
 
 The Rationale for Utilizing Information Derived from 
 Automated Systems in the Management Process 
 
 The fact is that the computer industry has been rather good at demon- 
 strating how various systems can make cataloging easier or typists type 
 faster, but they have been rather negligent in showing managers how they 
 can manage better. If one looks at many of the reasons why managers resist 
 automated systems, he will usually encounter a motivational problem. 
 Unfortunately either managers are motivated to use automated systems or 
 they are not. It is unfortunate because the systems designers have given 
 library managers little reason to be motivated. The purpose of the last 
 section of this paper is to propose a reason for managers to use automated 
 systems. 
 
 In an article, A.B. Cherns points out that automation can be used as a 
 management tool for either centralization of services, or for decentraliza- 
 tion of the organization. It can make centralization easier because it pro- 
 motes communication of necessary administrative information from the 
 line operations to the decision-makers in the organization. Thereby, the 
 administrators can make decisions that are timely and well informed. 
 Similarly, in a decentralized organization the problem tends to be that each 
 independent section of the organization has difficulty knowing precisely 
 what is expected of it. The improved communication capability of auto- 
 mation data can allow decentralized units to react to changing environ- 
 ments in a clear and uniform manner. 25 This paper will devote itself to 
 centralized organizational aspects of library automation. It will be address- 
 ing particularly those who are library administrators, or those who wish to 
 become administrators. 
 
 If one remembers the first section of this paper, the information flow 
 from level-to-level within the organization was reviewed. It was found that 
 one of the basic problems for the decision-maker in the organization was 
 getting reliable information concerning the needs of the clients. There are 
 those who would argue that this kind of problem is solved by instituting 
 multi-level committees or quality circles where the line personnel can 
 directly approach the manager with information. Unfortunately this still 
 does not solve the basic deficiency, which is that the decision-maker still 
 
106 JOHN N. OLSGAARD 
 
 does not know the actual extent of the problem or the accuracy of the 
 information. This is perhaps where the use of automation data has its best 
 use. 
 
 There are two essential characteristics of successful management: 
 superior information and superior control. 
 
 Superior Information 
 
 With superior information-gathering via automated systems, the 
 organization can progress from being a purely reactive entity (i.e., only 
 responding from crisis to crisis), to being an aggressive marketing organi- 
 zation. Thus, the organization can respond to situations before they reach 
 the critical stage, or a phase that is damaging to the credibility of the 
 organization. Information derived from automated systems does not just 
 produce quantitative data, but has the potential for a direct qualitative 
 effect on the library. This level of information produces better decisions, 
 and makes the organization more responsive to client needs. 
 
 Superior Control 
 
 There is a second reason that library managers should consider mak- 
 ing more use of data derived from automated systems. That rationale is that 
 with information from automated systems, the library manager can gain 
 more control and power over his organization. One somewhat hesitates 
 broaching the topic because speaking of "power" with today's organiza- 
 tions being geared to humanistic models is considered heretical, or at least 
 in bad taste. As a working definition I will define power as the ability to 
 influence change in another person or group of people. In 1 959 French and 
 Raven did a good deal of basic reasearch into the types of power that can be 
 exerted within a social organization. Their conclusion was that there are 
 basically five types of power: 
 
 1 . Reward power. The ability to reward an employee for correct action. 
 
 2. Coercive power. The ability to punish an employee for incorrect 
 behavior. 
 
 3. Legitimate power. The employee believes the employer has a right to 
 prescribe behavior. 
 
 4. Referent power. The ability of the employee to adjust behavior via 
 identification with the employer. 
 
 5. Expert power. The employee adjusts behavior because the employer has 
 some special knowledge. 26 
 
 In 1981 Yuki further delineated these power bases within organizations. 27 
 Up until about twenty years ago the library manager commanded 
 power through the first three kinds of power described earlier. The man- 
 ager was invested by the institution with legitimate power of position. 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 107 
 
 Some of the rights of that position included the almost total ability to 
 reward or punish the employees for which the manager was responsible. 
 The library, for better or worse, was in the hands of a single individual. 
 That was twenty years ago. 
 
 Since then, a number of factors have affected this power base of the 
 library manager. In probably one of the most cited pieces in library 
 literature, Downs and McAnally portrayed the problems of the academic 
 
 00 
 
 library manager. On one hand the academic administration was limiting 
 the resources and the privileges of the library manager, and on the other 
 hand the library faculty were gaining more individual rights of employ- 
 ment. Although the library manager is still invested by the institution with 
 the responsibility for effectively operating the organization, the institution 
 has taken over an increasing array of budgetary decisions. One need not 
 itemize the research on the loss of power of the library manager over library 
 employees. The literature is packed with the joys of dealing with faculty 
 status, collective bargaining and, of course, participative management. 
 The library manager finds himself with the same responsibilities as twenty 
 years ago, but without the ability to effect change either with the organiza- 
 tion as a whole or with individual employees. In the modern library 
 organization, rewards are given on the basis of union contract or commit- 
 tee judgment rather than by a decision by the library manager. Similarly, 
 the coercive power of the library manager has been delegated to legal 
 council and union steward. Many librarians are in one stage or another of 
 seeing this phenomenon occur in their organization. The traditional basis 
 for power over their organization is being eroded or is already gone. 
 
 What can the library manager do? The profession cannot throw out 
 the unions or the committees because they are here to stay. This paper is 
 also not suggesting that library managers give up trying to direct their 
 organizations. What it is suggesting is that library managers must find a 
 new basis for power to effect change. Since few of us are blessed with the 
 charisma to lead on the power of our personality, that new power base or 
 control must occur through superior information about the organization. 
 
 Salancik and Pfeffer have developed a model for gaining power within 
 an organization known as the "Strategic Contingencies Model." 29 The 
 foundation of this model is the fact that whoever controls the resources of 
 the organization, controls the organization. Part of the necessary resources 
 of an organization is information. This is where the library manager can 
 exert new power. He/she can determine that his/her office is the only 
 central collection point for management information and can determine 
 who, when and under what circumstances that information will be distrib- 
 uted. Thus the basis for power for the future library manager lies in the 
 area of expert power. The library manager will possess unique and vital 
 organizational data available to no one else. As Pfeffer and Salancik point 
 
108 JOHN N. OLSGAARD 
 
 out: "It is the case that if one controls the information used in decision 
 making, one can control the outcomes." 3 
 
 The manager can further exert control over employees by using, or 
 withholding, information. Meltzer has suggested that the psychological 
 need-to-know is a powerful motivating force in organizations. By with- 
 holding information the manager can actually inflict punitive control 
 over a given employee. 
 
 The requirement for this kind of power rests in the ability to collect 
 the information. This is where data gathering from automated systems is 
 absolutely vital. Under present capabilities almost every aspect of a library 
 operation can have an automated reporting function. As such, every 
 library operation can have day-to-day practically individual-by- 
 individual direct reporting to the library manager. There is no need for 
 the constant filtering and refiltering of information presently available. 
 
 There are those who may think that this is too much like a chapter 
 from 1984 or a sequence Irom2001: A Space Odyssey. But like many things, 
 data gathered from automated systems is merely a tool. How library 
 managers use this tool is largely a matter of personal discretion. However, 
 the use of automation as a vehicle to effect change within the library is one 
 of the last opportunities for library managers to exert any kind of control 
 over the direction of their organization. 
 
 Conclusion 
 
 First, this paper examined the library as a marketing entity and 
 discovered that most libraries are internally-oriented organizations. It was 
 proposed that this orientation was due to the inability of libraries to 
 adequately change with the differing needs of its clients. Second, the causes 
 of nonuse of data from automated systems by library managers for 
 decision-making were examined. It was postulated that this nonuse was 
 primarily due to a motivational failure on the part of library managers. 
 Last, the reasons why library managers should utilize data derived from 
 automated systems, both as a way of directing the entire organization to a 
 more client-oriented position, and as a means for the library manager to 
 gain more personal control over the library, were examined. 
 
 In the final analysis, experience and instinct will ultimately mark the 
 best library managers. However, good decisions are not only based on 
 instinct, but also on the ability to formulate that instinct around quality 
 information concerning the changing environment we live in. Rose has 
 pointed out that the modern manager needs to be cybernetic as well as 
 literate and numerate. 2 The responsibility of the library manager in 
 today's society is more complex and difficult than ever before. While the 
 demands on the services of the library are greater, the tools that the 
 
CHARACTERISTICS OF MANAGERIAL RESISTANCE 109 
 
 manager can use to meet those demands are fewer in number and more 
 limited in scope. One of these tools is the kind of quality information that 
 can be derived from automated systems. In order to carry out the responsi- 
 bility, to do the job he or she is being paid to do, the library manager cannot 
 afford not to use the tools available, including administrative information 
 from automated systems. 
 
 REFERENCES 
 
 1 . Runyon, Robert S. "Towards the Development of a Library Management 
 Information System." College & Research Libraries 42(Nov. 1981):539-48. 
 
 2. Drake, Miriam A. "Managing Innovation in Academic Libraries." College & 
 Research Libraries 40(Nov. 1979):504. 
 
 3. Miller, James G. "A Theoretical Review of Individual and Group Psychological 
 Reactions to Stress." In The Threat of Impending Disaster, edited by George H. Grosser, et 
 al., pp. 13-14. Cambridge, Mass.: MIT Press, 1964. 
 
 4. Meier, Richard L. "Information Input Overload: Features of Growth in 
 Communications-Oriented Institutions." In Mathematical Explorations in Behavioral 
 Science, edited by Fred Massarik and Philburn Ratoosh, pp. 233-73. Homewood, 111.: Dorsey 
 Press, 1965. 
 
 5. Horn, Roger. "The Idea of Academic Library Management." College b Research 
 Libraries 36(Nov. 1975):464-72. 
 
 6. Andreasen, Alan R. "Nonprofits: Check Your Attention to Customers." Harvard 
 Business Review, 60(May/June 1982):105-10. 
 
 7. Meltzer, Morion F. Information: The Ultimate Management Resource. New York: 
 AMACOM, 1981, p. 6. 
 
 8. Shera, Jesse H. The Foundations of Education for Librarianship. New York: 
 Becker & Hayes, 1972, pp. 394-95. 
 
 9. Divilbiss, J.L. "Problems of Teaching Library Automation." In Problems and 
 Failures in Library Automation (Proceedings of the 1978 Clinic on Library Applications of 
 Data Processing), edited by F. Wilfrid Lancaster, pp. 67-74. Urbana-Champaign: University 
 of Illinois Graduate School of Library Science, 1979. 
 
 10. Toffler, Alvin. Future Shock. New York: Bantam, 1970, pp. 350-58. 
 
 1 1 . Firnberg, David. Computers, Management and Information. London: George Allen 
 and Unwin, Ltd., 1973, p. 20. 
 
 12. Posner, Michael I. "Components of Skilled Performance." Science 152(24 June 
 1966): 171 2-1 8; and Miller, George A. "The Magical Number Seven, Plus or MinusTwo: Some 
 Limits on Our Capacity for Processing Information." Psychological Review 63( March 
 1 956 ):8 1-97. 
 
 13. Melnyk, Vera. "Man-Machine Interface: Frustration." Journal of the ASIS 
 23(Nov./Dec. 1972):392-401. 
 
 14. Lacey, Paul A. "Views of a Luddite." College b Research Libraries 43(March 
 1982):! 10-18. 
 
 15. Ennis, Philip H. "Technological Change and the Professions: Neither Luddite Nor 
 Technocrat." Library Quarterly 32(July 1962):189-98. 
 
 16. "How to Conquer Fear of Computers." Business Week, no. 2731 ,29 March 1982, pp. 
 176-78. 
 
 17. Rose, Michael. Computers, Managers, and Society. Harmondsworth, Eng.: 
 Penguin, 1969, pp. 206-11. 
 
 18. De Greene, Kenyon B. "Technological Change and Manpower Resources: A Systems 
 Perspective." Human Factors 17(Feb. 1975):52-70. 
 
 19. Sanders, Donald H. Computers and Management. New York: McGraw-Hill, 
 1970, p. 327. 
 
110 JOHN N. OLSGAARD 
 
 20. Easton, Allan. Complex Managerial Decisions Involving Multiple Objectives. 
 New York: John Wiley & Sons, 1973, pp. viii-x. 
 
 21. White, Lynn Townsend. Machina Ex Deo: Essays in the Dynamism of Western 
 Culture. Cambridge, Mass.: MIT Press, 1968, pp. 57-73. 
 
 22. Donaldson, Hamish, et al. Computer by the Tail: A User's Guide to Computer 
 Management. London: George Allen and Unwin, Ltd., 1976, pp. 45-46. 
 
 23. Montague, Eleanor. "Automation and the Library Administrator." Journal of 
 Library Automation ll(Dec. 1978):317. 
 
 24. Orlicky, Joseph. The Successful Computer System: Its Planning, Development, and 
 Management in a Business Enterprise. New York: McGraw-Hill, 1969, p. 4. 
 
 25. Cherns, A.B. "Speculations on the Social Effects of New Microelectronics 
 Technology." International Labour Review 1 19(Nov./Dec. 1980):705-21. 
 
 26. French, John R., Jr., and Raven, Bertram. "The Bases of Social Power. "In Studies in 
 Social Power, edited by Dorwin Cartwright, pp. 155-65. Ann Arbor: University of Michigan, 
 1959. 
 
 27. Yuki, Gary A. Leadership in Organizations. Englewood Cliffs, N.J.: Prentice-Hall, 
 1981,38-58. 
 
 28. McAnally, Arthur M., and Downs, Robert B. "The Changing Role of Directors of 
 University Libraries." College b Research Libraries 34(March 1973):103-25. 
 
 29. Pfeffer, Jeffrey, and Salancik, Gerald R. The External Control of Organizations. 
 New York: Harper fe Row, 1978. 
 
 30. Ibid., p. 270. 
 
 31. Meltzer, Information, pp. 15-16. 
 
 32. Rose, Computers, Managers, and Society, p. 10. 
 
MARY ELLEN JACOB 
 
 Director for Library Planning 
 OCLC 
 
 NEAL K. KASKE 
 
 Director 
 OCLC Office of Research 
 
 Management Information Systems 
 in a Network Environment 
 
 What do the terms management, information and system imply? Manage- 
 ment implies control, monitoring and some type of role. Information is 
 more than data, and it is not always knowledge it can lead to knowledge. 
 The term system implies organization, order and plan. 
 
 In looking at a management information system (MIS), a system 
 which supplies management information or information to management, 
 we need to look first at the function of management. Peter Drucker has 
 given a number of definitions on both management and the role of manag- 
 ers. One of these is: "Management exists only in contemplation of perfor- 
 mance." 1 This suggests that management is not an end in itself but a means 
 to an end, and that same aspect of it applies to management information 
 systems. 
 
 Since we are librarians trained in engineering, the dual approach of a 
 technologist and a humanist seems appropriate. Thus, combining system 
 analysis with an assist from Rudyard Kipling's "Six honest serving men," 2 
 we should examine the questions, what, why, when, how, where, and who. 
 We at OCLC need to look at management information from several 
 different perspectives, which we will discuss more fully under "who." But 
 much of what we collect and provide must be from the library manager's 
 perspective. 
 
 What 
 
 MIS's focus is primarily on allowing us to do a better job as 
 managers not to create a system for acquiring information. While this 
 point has been made by others, it deserves emphasis, for too often its 
 
772 MARY ELLEN JACOB b NEAL K. KASKE 
 
 meaning is lost in the masses of data we can accumulate through systems, 
 particularly through automated systems. In the OCLC network we process 
 approximately 2 million messages per day, generating six to eight reels of 
 tape or equivalent disc files. 
 
 Our problem with automated systems is not a lack of data, but how to 
 sift through the mass of data available to find useful information, how to 
 translate that information into knowledge that will enable us to make 
 better decisions, and how to better plan, monitor and control the use, 
 growth and development of the systems. 
 
 We need to segregate those things that we should examine in detail 
 from those things we can look at in aggregate. There is really little point in 
 reviewing transactions that fall within the norms. On the other hand, for 
 those transactions that lie outside the norms, we may need to look at 
 detailed information to ascertain the reasons for their abnormality. For 
 example, if it appears that all books on an order which take longer than 
 nine weeks to receive are books coming from overseas, we may want to add 
 some different parameters that allow extra time for such books. Maybe the 
 only books we want to select are ones that take longer than twelve weeks 
 (i.e., we may need different norms for different classes of materials). The 
 same idea can apply to cataloging of materials. Different formats and/or 
 different subject areas may require different amounts of time and effort. 
 
 We can ask many questions of management information systems, and 
 we can get many answers, but are we asking the right questions? What do 
 we need to know to improve the operation of our libraries, and for whom 
 the library manager, library staff or user? 
 
 Too often the tendency has been to focus on suboptimization (e.g., to 
 make acquisitions or the cataloging process happen in the fastest possible 
 manner). More recently, technical services departments have sought to 
 ensure that materials move through the technical services area in toto in 
 the fastest possible way, rather than moving them through quickly by 
 individual subunits. 
 
 We really need to look at the overall library operation and not the 
 individual parts. The process, from the request for materials (if our pro- 
 curement is triggered on that basis) to the actual provision of the material 
 to the end user, may be one parameter. We may be trying to provide 
 materials in advance of requests. In that case we may want to see how 
 successful our selection criteria has been. How many times has a particular 
 item been used that we have recently acquired? If it has not been used at all, 
 what are the reasons for its nonuse? Did we select correctly or not? How 
 does this relate, not only to our current users, but to future users? These are 
 all questions that must be answered, particularly in these times of limited 
 budgets for materials and staff. 
 
MIS: NETWORK ENVIRONMENT 113 
 
 Why 
 
 Having discussed some aspects of what, we should also consider why 
 we need information. What are our criteria for success? Do we measure 
 them based on the number of books processed, patrons served or user 
 satisfaction? We have few measures of satisfaction other than our use of 
 statistics or surveys relating to users and nonusers. The latter information 
 is difficult, costly and time-consuming to acquire. However, if the library 
 were to become a community information source accessible to every home 
 in that community, we could acquire considerably more data easier, and in 
 a more timely manner. Mr. Dowlin has described one system with this 
 potential Maggie's Place elsewhere in these proceedings. 
 
 For a number of years, and particularly in recent planning, public 
 libraries have been focusing on user satisfaction as a major criterion. 
 Considering the political process in which libraries are involved with 
 taxpayer support, this may be an appropriate measure of usefulness fora 
 public library. 
 
 Academic institutions have always felt that their facilities were estab- 
 lished for current users as well as for future users; they were equally 
 concerned about building a scholarly collection for both the present and 
 the future. It is more difficult to anticipate and establish objective measures 
 of how well the library meets that future use. We can only do it by looking 
 at how well we have met the needs of scholars in building our retrospective 
 collections, and can only hope to guess correctly about the future. 
 
 When 
 
 When we need information can often be as critical as what informa- 
 tion, and why we need that information. Timing is, in many instances, 
 determined by the nature of the process or by cycles within our parent 
 institutions, such as a calendar or fiscal year. We have more control over 
 information generated for our internal management needs. We often fall 
 prey to the belief that we need instantaneous access to current information. 
 Computer systems offer us two major advantages. They offer us virtually 
 instantaneous access to much information, as well as access to vast 
 amounts of data that can be readily analyzed in many ways. 
 
 Computers can offer us information in real-time, but we need to think 
 carefully about that use. There are times when delays in our systems work 
 to our advantage. In designing systems we need to consider how to provide 
 that kind of tolerance. We may at times feel something is abnormal in 
 relation to a present situation but, if examined later on a long-term basis, 
 may have been a minor blip no major problem. We need to build certain 
 tolerance levels into our systems to provide this aspect, so we take action 
 when needed, but not unnecessarily. 
 
114 MARY ELLEN JACOB b NEAL K. KASKE 
 
 We also need to recognize that much of our data is historical i.e., it is 
 based on our current systems and technology and our present methods of 
 use. When we use historical data to predict the future, we must recognize 
 that they will predict the future only insofar as the future is like the past. 
 But if conditions occur where significant changes may happen, we need to 
 explicitly acknowledge such changes and estimate the impact they may 
 have on our history-based projections. 
 
 This is one of the major challenges we face in designing online 
 catalogs. Since much of our information is derived from catalog use 
 studies, we must not extrapolate the limitations of card catalog access into 
 a new tool (the online catalog). We need to separate those aspects of 
 information-seeking which are not dependent on the tools used but are 
 inherent in the process from those things which are really based on the 
 form or medium used. 
 
 How 
 
 The focus of this clinic is primarily on management information in 
 automated systems. Used carelessly, computer systems can increase our 
 information overload so that we have too much information to make the 
 decisions we need to make. Rather than reducing our risk, these systems 
 increase our confusion and literally make decision-making impossible. On 
 the other hand, if we use those systems to create normal patterns and ignore 
 what falls within the norms (i.e., have the system select the items we should 
 look at), we can reduce the volume of information and concentrate on 
 matters that need our attention. 
 
 Computers have a fantastic ability to process large amounts of infor- 
 mation rapidly and whittle the information down to important items of 
 concern to us. To date, they do not have the ability, however, to make 
 decisions based on that information. They can merely alert us to the fact 
 that some kind of action should be considered. 
 
 In data collection we must focus on automatic means of collecting 
 data which are a normal part of the process for a particular activity. It 
 should not require separate actions or unique actions just to generate that 
 data. If it does, we run two risks: ( 1 ) someone will forget to take the action 
 to collect the data, or perhaps more likely (2) we will add to the cost of our 
 overall processing by collecting such data. Information has a cost. We must 
 ensure that the cost of collection and analysis does not outweigh the 
 usefulness of the information. 
 
MIS: NETWORK ENVIRONMENT 115 
 
 Where 
 
 Management information is pervasive and affects all parts of our 
 operation. We were asked to look at this topic within the context of the 
 network environment. Consequently, we will limit our remarks mainly to 
 this environment. 
 
 At OCLC much of our early work focused on acquiring only those 
 pieces of data which we actually needed to run and support the operation 
 in particular, the information required for billing. The design of our 
 systems from a management support perspective leaves much to be desired. 
 We are trying to rectify that in the design and implementation of system 
 enhancements, but more importantly in the design of new systems. 
 
 For example, a major function to be added within the OCLC Interli- 
 brary Loan Subsystem version three is the statistics-generating capability 
 that will allow users to document the kinds of activity that take place 
 within the interlibrary loan operation for both borrowers and lenders. 
 Clearly, this is only the tip of the iceberg. Much more information is 
 available and could be collected, but it represents consensus among users of 
 the Interlibrary Loan Subsystem for those items they consider to be impor- 
 tant. One can identify similar needs within acquisitions, cataloging and 
 serials control. 
 
 Another aspect of "where" relates to whether such analysis and collec- 
 tion are performed online or offline. Not all processing needs to be online. 
 Where quick, short, unique answers are needed, online has advantages 
 from both a management and systems perspective. If long reports are being 
 generated, these are better done offline. Much depends on the manager's 
 needs and time frame and the system's design and flexibility. 
 
 Often what is needed is not merely to see a library's performance in 
 isolation, but to see that performance over a period of time, preferably in 
 comparison with other libraries. The OCLC system has the capability for 
 providing such comparisons. 
 
 Some time has been spent by OCLC staff in designing, in a broad way, 
 a MIS. Whether the system will ever exist depends on the needs of our users, 
 the priority they give management information as opposed to additional 
 indexing enhancements, and new features for other subsystems including 
 subject access. At the moment, management information seems to be a low 
 priority. Thus, while it is highly desirable and could be useful to library 
 managers, other things are more important. 
 
 We are doing a much better job of designing management informa- 
 tion into the local library system from the beginning rather than as an 
 afterthought. We have reviewed what libraries need to know in regard to 
 local library functions, and how we can provide this information in a 
 cost-effective manner. Libraries will be able to collect statistics on just 
 
116 MARY ELLEN JACOB b NEAL K. KASKE 
 
 about everything. The need, however, is to determine the key factors in 
 terms of library performance, and to collect and analyze that data rather 
 than collecting everything. This will only answer one of the questions I 
 have posed, namely, how the library relates to itself, and will not give 
 answers about how the library relates to other libraries. That information 
 must originate outside the institution, and could be provided through the 
 mechanism of the central system. Some work has been done in this area 
 (e.g., the Council for the Advancement of Small Colleges has used the 
 HEGIS data), and using input data from a library will provide informa- 
 tion to the library on its own processing in relation to other libraries. 3 
 
 Who 
 
 In looking at information from a network perspective, we need to 
 consider the different types of people who require and use management 
 information. Our preceding discussion provides the necessary back- 
 ground. We have the end user or the operator at a terminal. We have library 
 management, regional network management or service representative. We 
 have the network itself or the service provider. Within each of these groups 
 are subgroups which need other types and levels of information. 
 
 The focus of management information systems has been, and con- 
 tinues to be, the manager. While that is important, our focus should be on 
 how to achieve the kind of performance we consider desirable. If we were 
 doing this, our focus would be on the individual operator and worker 
 rather than on the manager. This change of emphasis is most important 
 within automated systems. 
 
 Operators need instant feedback on what they are doing right and 
 what they are doing wrong. Sometimes this feedback is merely the 
 response "The system did not understand your last command" or it 
 may be more complex, depending on the nature of the system. The opera- 
 tor should be able to ascertain performance measures for each session 
 completed. Few systems presently provide these performance measures. 
 This is the kind of information we should be designing into future systems. 
 
 Users of online systems want to know: How can I get the information I 
 want most easily? What am I doing wrong? How can I fix those things that 
 are wrong? Library staff are interested not only in those facts, but also: 
 How long did I take? How much did I do? How well did I relate to what 
 others have done? 
 
 Library managers may be interested in parameters associated with 
 individual operators if they are immediate supervisors. Senior manage- 
 ment is more interested in how the library is doing in relation to its past 
 performance, trends and growth in activity, how the library performance 
 relates to other libraries, and what are the means of focusing on those areas 
 
MIS: NETWORK ENVIRONMENT 117 
 
 which deviate from the normal performance range so as to take whatever 
 corrective action is necessary. 
 
 In addition, senior management is also interested in assessing "what 
 if" situations. "If my budget were reduced 10 percent, how would that 
 affect resources available for staff, serials, monographs, and other library 
 activities?" Would it be better to take a 10 percent cut in staffing, in 
 materials area or different percentages of cuts in each? Should one area be 
 favored over another? Historical data and models can help the manager 
 assess the impact these changes might have on the library and its operation. 
 They are not a substitute for intelligent judgment, and one must always 
 look at the underlying assumptions, particularly when using historical 
 data. 
 
 Regional network managers are interested in matters very similar to 
 those of the library manager e.g., the past versus the present; trends versus 
 growth; this network versus other networks; high and low usage, and the 
 reasons behind such usage; errors and problems incurred by users with the 
 system; and the "what if" situation. 
 
 Bibliographic cooperative management is similarly interested, but 
 from a somewhat different perspective. They are interested in present 
 activity versus past, in trends and growth, particularly in relation to system 
 performance and future capacity planning. They are less interested in their 
 performance versus other cooperatives, but are very interested in how 
 different groups of libraries are performing in relation to each other and as 
 a whole. They are interested in high and low usage, and reasons for such 
 usage. They also have an interest in error prevention and operator prob- 
 lems. Again, models and "what ifs" can be used to pinpoint potential 
 problems, their impact on capital equipment procurement decisions, and 
 the timing of those decisions. While each level is not necessarily interested 
 in the same information, they are still operating from a common base (i.e., 
 system activity), and this relates back to how an individual operator uses 
 and responds to the system. 
 
 If we give the terminal operator or user the kind of information needed 
 for the activity being performed, it should be somewhat easier to pass along 
 the information the manager needs, and so on up the chain of command. 
 The important thing is to select for each level of use the items that matter at 
 that particular level, and not to prepare a mere mass of data. It is to select 
 information that helps in the decision-making process as in the case of 
 the operator to improve his/her performance. In the case of the manager, 
 to plan, monitor, control, and alter the operation of the overall system. 
 
118 MARY ELLEN JACOB b NEAL K. KASKE 
 
 Research Activities 
 
 In moving from management information systems to the individual 
 users, we will bring together two research and development activities 
 taking place in libraries today. These research and development activities 
 are those of online cataloging and library MIS. We will start by reviewing 
 some of the initial findings, assumptions and research activities relating to 
 both online catalogs and MIS. 
 
 Before discussing these areas, it is important to note the connection 
 between management information systems and online catalogs or OPACs 
 (Online Public Access Catalog systems). The connection between these 
 two areas is that of cause and effect. Used together, they create an atmo- 
 sphere for "user studies deluxe! " The record of use that is possible to create 
 from an online catalog provides a library's management information 
 system with accurate data on how the catalogs are being used by patrons. 
 Catalogs are the key to library collections, and an understanding of their 
 use would be a key to understanding library use. These online catalogs, 
 however, have a number of barriers. 
 
 The initial findings discussed earlier come from a number of different 
 studies. All of these studies have been conducted in part or in their entirety 
 by the Office of Research at OCLC. These studies are on three major topics: 
 subject access, terminal requirements (queueing), and the online catalog. 
 
 Subject Access Project 
 
 The subject access project's objective was to determine the features of 
 an automated subject retrieval system that would support the present 
 search tactics employed by library users. The final report for the project 
 will be issued this fall as part of the OCLC's Office of Research "Research 
 Report Series." This report, like the first one originating from this project, 
 will be authored by research scientist, Karen Markey. 4 Two other papers on 
 the subject have also been published during the course of this project. 5 
 
 Terminal Requirements Project 
 
 The second study, "Terminal Requirements for Online Catalogs in 
 Libraries," is being conducted with funds from the National Science 
 Foundation. Its purpose is to develop and to test an algorithm for estimat- 
 ing the number of public computer terminals needed by a library to 
 support an online public catalog. The results of this study will be pub- 
 lished in the literature, and the guidelines for the use of the algorithm will 
 be offered as part of OCLC's monograph series (Library, Information, and 
 Computer Science Series). This work, coauthored by Neal K. Kaske and 
 John Tolle, will be available in 1983. 
 
MIS: NETWORK ENVIRONMENT 119 
 
 Online Catalog Project 
 
 The last of the three study areas is devoted to the online catalog and 
 consists of three major activities. The first one is reported by Kaske and 
 Ferguson in a report issued by the Council on Library Resources, Inc. 
 (CLR), the major funder of these three projects. 6 The second activity, 
 "Online Public Access Systems: Data Collection Instruments for Patron 
 and System Evaluation," which has just been completed, was composed of 
 three basic phases. The first phase was to assist in the development of 
 patron assessment tools. Phase two was to examine and compare several 
 online patron access systems. The third phase recommended a uniform 
 online catalog patron monitoring technique. Most of this report (phase 
 two) has been published separately as the monograph Online Public 
 Access Catalogs: The User Interface, by Charles R. Hildreth 8 and is availa- 
 ble from OCLC. 
 
 The third activity is the Online Catalog Research Project funded by 
 CLR and titled, "Online Public Access Systems: Data Collection and 
 Analysis." The ultimate goal of this research is to improve, through the 
 design and enhancement of online catalogs, the patron's ability to access 
 information. The project is divided into three major phases. The first 
 phase is data collection and analysis via patron questionnaires and 
 focused-group interviews. The second phase analyzes the current patterns 
 of use made of online catalogs via transaction logs and activity reports. The 
 final phase evaluates and integrates the findings for library management 
 decision-makers. 
 
 Initial Findings 
 
 An initial key finding from these research projects was that there are a 
 number of barriers which prevent library patrons from effectively using (or 
 in some cases from even using) online catalogs. The barriers are the 
 computers themselves, the system's language, and the bibliographical 
 information. 
 
 The first barrier is computerization itself. Many people believe com- 
 puters to be complex and that formal training is necessary to use computer- 
 based systems. Some fear that they may "break" the computer, or 
 inadvertently cause the computer system to malfunction if they do not use 
 it correctly. People also find the computer to be dehumanizing and, as a 
 result, do not want to learn about it. Some say that computers put individu- 
 als out of work, and therefore, should not be used. Thus, the computer (or 
 terminal) and how it operates is a basic barrier to its use. 
 
 Patrons remember using the card catalog as an old familiar friend. 
 There are many people who prefer the card catalog because it is more 
 private i.e., others cannot easily see what one is searching. It isalso easier 
 
120 MARY ELLEN JACOB to NEAL K. KASKE 
 
 to look scholarly while intently scanning a series of catalog card en tries. In 
 contrast, the online catalog is generally in a public area where others may 
 easily view a search in progress. This may have significant implications for 
 where terminals are located and how many are required. 
 
 Another barrier is that the operation of various terminals is somewhat 
 different. One of the most frustrating things for individuals to overcome in 
 first learning how to use a computer terminal is to remember to depress the 
 return key so the computer will read the message that they have keyed in. 
 There are other idiosyncratic characteristics about computer terminals, 
 many of which become trivial when one learns how to operate them. But 
 until a few basic operations have been learned, terminals may be as foreign 
 to a user as a manual transmission is to one who has driven only an 
 automatic transmission. 
 
 Another major barrier is the system's or command language. Here 
 we must learn how to ask the computer a question and get an answer. Is it 
 an A VI and the first four letters of the author's last name, or is it something 
 even more bizarre and complex such as A / T with the first four characters of 
 the author's last name and the first four characters of the first significant 
 word of the title? 
 
 Some systems have two levels of command languages: one for the 
 novice which none of us admit to being and another for the expert a 
 level we have not yet achieved. How about that vast majority in the middle? 
 To these the system acts as a barrier. 
 
 The third barrier is bibliographical information itself. It seems that 
 no matter how often people are given lectures on bibliographical informa- 
 tion, they arrive at universities and public libraries with either near full 
 knowledge or no knowledge at all of how to interpret bibliographical 
 information displayed on a 3 by 5 inch card, a COM (computer output 
 microfilm) catalog, or information on a computer terminal. A lesson to 
 learn here is that major elements of bibliographic records should be clearly 
 labeled with words easily understood by the majority of users. 
 
 The barriers then are computers, systems language and bibliographi- 
 cal records. The first barrier will soon disappear as computer users become 
 more numerous, be it via the home computer or classroom instruction. We 
 should see real improvements in the second as we make system languages 
 more understandable and "user friendly." The third barrier, bibliographi- 
 cal records, remains to be addressed. 
 
 Another initial finding is that while information needs are not all 
 ilike, there are two key elements. These elements are time available to 
 research a topic and knowledge of the subject. Most informational needs 
 have an attached time constraint. That time constraint may be only a 
 matter of minutes or hours, or it may be open ended in the sense that it will 
 last for years. An example of an informational need that must be satisfied 
 
MIS: NETWORK ENVIRONMENT 121 
 
 in hours or possibly minutes is one expressed by the student who walks up 
 to the reference desk and says: "I've go to give this speech next period on 
 birth control. What do you have in the library on birth control? Or by a 
 student who says: I have to have an essay ready by next period on the subject 
 of the Falkland Islands. Have you ever heard of the Falkland Islands and, if 
 so, where can I find information? 
 
 Other informational needs may develop over years. An expert on herbs 
 and herb gardening, when traveling about the country on vacation or 
 business, may take advantage of different libraries and examine their 
 collections as to holdings. The individual will probably also make an 
 effort to visit different herbariums. As the knowledge grows, the informa- 
 tional needs change indicating that informational needs are time related. 
 
 An individual's subject knowledge is important in satisfying any 
 given informational need. The novice in a field needs a great deal of 
 information e.g., definitions and basic works. The expert, on the other 
 hand, needs to find that last elusive informational package on a subject, 
 held by who knows what library; and the information should be current 
 perhaps from the latest journals or other information resources. 
 
 Assumptions 
 
 With this knowledge that people's informational needs are time 
 related, and that they are at different levels of knowledge on given subjects, 
 we need to make some assumptions. We must start by reviewing previous 
 catalog studies though they will be about card catalogs, book catalogs or 
 COM catalogs not about online catalogs. More important than the fact 
 that the catalog medium is different is that previous catalog studies were 
 only time slices; they were not longitudinal studies. Consequently, earlier 
 studies have limited application to the design and use of online catalogs. 
 
 People do not perform "known item searches" for most of their 
 informational needs. They do all types of searches in more than one library 
 location (the college library, the public library, etc.) to satisfy their infor- 
 mational needs. Figure 1 shows that as the patron moves "across time" in 
 different library locations, they will do many kinds of searches for informa- 
 tion. For example, in the first library they may search under author, then 
 under title, then under subject. They may also be knowledgeable enough to 
 use titles as subjects, depending on the division of the catalog. If it is a 
 dictionary catalog, this is done many times unwittingly. If a catalog is 
 divided, a person looking for a subject item and only starting with the 
 subject may use the title catalog or the author/ title catalog as a subject 
 catalog, hoping the first significant word in the title, minus the articles, 
 will be the subject they are looking for. They then find a book with that 
 kind of title, note the Library of Congress subject heading on the card and 
 
722 
 
 MARY ELLEN JACOB b NEAL K. KASKE 
 
 promptly switch to a subject catalog to find their subject material. If their 
 information need is more than a few hours or a few days, people may move 
 on to another library to search its holdings in a similar way. They may 
 continue at a third and fourth library, again searching by author, title, 
 subject, and by title as subject. To gain a clear picture of how people 
 conduct subject searches, longitudinal time studies need to be performed. 
 
 Type of 
 Searches 
 
 Author 
 
 Title 
 
 Subject 
 
 Title as 
 Subject 
 
 Library 1 
 
 Library Locations 
 Library 2 Library 3 Library 4 
 
 c 
 
 tz 
 
 T1me 
 
 2401 
 
 Fig. 1. Patron Searching Pattern Overtime 
 
 Transaction Analysis 
 
 Having examined some initial findings and having looked closely at 
 our underlying assumptions, we should be able to provide some answers to 
 our research questions How are online catalogs being used? An overview 
 of the related projects has been reported earlier. How do we obtain answers 
 to these questions? Our solution is transaction analysis (TA). 
 
 The kind of TA we are going to discuss is not transactional analysis of 
 the popular sense, but it is transaction analysis. The kind of TA we are 
 discussing is the type that makes use of the transactions log produced by 
 computers. This log is a record of the computer dialogue between the user 
 and itself. These dialogues are recorded so that if computers fail, the 
 transactions log can be used to recreate the processes, and by replaying 
 them, bring the system back up from a "crash." This is done so people can 
 continue to work as they were before the computers failed. The data 
 elements that need to be captured so we can study them are listed and 
 briefly explained in the following: 
 
MIS: NETWORK ENVIRONMENT 123 
 
 1 . Session identifier. The unique identifier associated with the particular 
 session being monitored. 
 
 2. User identifier. The unique identifier associated with the individual 
 user being monitored. 
 
 3. Database or file being used. If relevant, the name of the file currently 
 being accessed by the user in a given database. 
 
 4. Date. 
 
 5. Time stamp. The time at which each transaction occurs. Time stamps 
 should provide as much accuracy as possible, although a time stamp 
 resolution exceeding hundredths of a second is not generally useful. 
 The point at which the time stamp will be applied must be specified. 
 Ideally, the input time stamp should be applied when the user com- 
 pletes the input (e.g., depresses the Enter, Return or other special 
 function key), and the output time stamp should be applied when the 
 first character of output is delivered to the user. Since these exact times 
 are not often available how the time stamps differ from the ideal time 
 stamps should be stated. 
 
 6. The source of each transaction. Possible sources should include at least 
 the terminal user, system and other transaction source (e.g., stored 
 command files or operator messages). 
 
 7. System-dependent state information. If other information about the 
 transaction is readily available, it should be included. Common 
 examples include a transaction code generated to govern internal 
 processing or special error or return codes. 
 
 8. Blank space. This is needed for state code assignment during post- 
 session analysis. 
 
 9. Length of text portion. Number of characters in the input or response. 
 10. Text portion. Contains the text of the user input or the system response. 
 
 The complete text is preferred when practical, but it may be truncated. 
 
 Why do we want to perform transaction analysis? The answers are 
 simple: to have better online services, to provide better access to libraries, to 
 make libraries respond to the informational needs of the public who uses 
 them, to provide the dream "a library at your fingertips" and foremost, 
 to make libraries the people's choice for information seeking. 
 
 Summary 
 
 In this discussion we have considered from our perspective how we 
 define management information systems in the network environment and 
 have considered Rudyard Kipling's "Six honest serving-men." 9 We have 
 reviewed general concerns that apply to any management information 
 system, but have also looked at this from the network perspective. We have 
 
124 MARY ELLEN JACOB b NEAL K. KASKE 
 
 focused heavily on users and their needs since they are the primary reason 
 libraries exist. We have looked at the patron's use of online catalogs; have 
 considered the barriers to catalog use, including computer knowledge, 
 systems and command languages; and, in the organization, the content of 
 bibliographic information. We considered informational needs as chang- 
 ing over time and with experience level. We have examined the impact of 
 library location on the search type, and we've looked briefly at transac- 
 tional analysis of the kinds of information that can be collected and some 
 of the analyses possible. 
 
 Three other issues we will touch upon but not fully deal with are 
 integration, privacy and power. We obtain management information not 
 from one system, but from many external, as well as internal, sources. 
 Presently we have left the integration of this information to the individu- 
 als. This lack of integration was a driving force behind OCLC's concern 
 for designing a management information system that uses not only net- 
 work and institutional information generated on the network system, but 
 also allows the manager to draw on other information sources such as 
 HEGIS, Bowker, census data, etc. In addition, it allows entry of other 
 information. At present this is a dream, but it can become a reality provided 
 librarians and managers believe such a system is useful and satisfies their 
 needs. Without user support, it remains simply an interesting concept. 
 
 Privacy issues are many and varied, and have been discussed at some 
 length in the library literature as well as in more general areas. There has 
 been considerable concern over the problems of computers i.e., their 
 ability to amass and analyze information related to individuals. There have 
 been a number of laws passed overseas and also within the United States 
 relating to the individual's right of access to such information to ensure its 
 correctness and validity. We must also consider these issues in designing 
 MIS for any kind of monitoring of a user's interaction with a computer- 
 based system. These issues are critical when we seek to monitor the activi- 
 ties and behavior of individual library staff members. How do we relate our 
 management functions? And what right of protection or redress does the 
 individual have in relation to such information? Management informa- 
 tion systems represent power in this sense, and we need to ensure that 
 proper consideration is given to the individual and his rights to privacy. 
 
 This, of course, leads us to the whole question of power within 
 management information systems. As was indicated by Mr. Olsgaard, there 
 is potential within management information systems for centralizing and 
 controlling information, for restricting access to it, and, consequently, for 
 it to be a source of power. While that is possible, it is also possible that we 
 will see a democratization of information rather than a concentration and 
 restriction of information as we are developing more computer-based 
 systems, particularly microbased systems which are linked to a central 
 
MIS: NETWORK ENVIRONMENT 125 
 
 database but which may act independently and maintain their own sepa- 
 rate stores of information. That information will be more readily available 
 to others who in turn can manipulate, access, analyze, reformat, and draw 
 conclusions from such information. 
 
 Earl Joseph, a futurist from Sperry Rand, has stated that we have 
 considerable difficulties with the development of distributive systems 
 because we have no models. Our religions and our culture all tend toward 
 hierarchical structures. We are not used to systems in which all parts are 
 equal. The U.S. Government represents a departure from democracy in a 
 political way. So too, distributed systems represent a new approach to 
 systems design and to the potential for democratization of system access to 
 and use. 
 
 We now face considerable challenges: how can we provide the integra- 
 tion of the systems we require and the information sources we need? How 
 can we provide privacy for the individual with relation to his/her rights 
 and privileges? How can we use power effectively not to restrict informa- 
 tion, but to make our systems work better and more effectively? 
 
 So, what do we see for the future? We will have the option of creating, 
 internally, our own systems which operate on microcomputers or mini- 
 computers. At OCLC we have used not only mainframe data collection and 
 analysis, but also such microbased tools as Visicalc and Supercalc for both 
 our system performance projections and preliminary budget analysis. 
 
 We will have access to information sources through such services as 
 EDUNET and EDUCOM, including the HEGIS data. We will likely have 
 library management-related services available from other information pro- 
 viders such as BRS and Lockheed, as well as OCLC and RLIN. Our 
 challenge as managers will be to take those various information sources 
 and integrate them so as to provide us with the information we need to 
 manage our individual enterprises. The questions remain to be answered. 
 How can we better manage our networks, our consortia, our libraries, to 
 better serve our users both present and future? 
 
 REFERENCES 
 
 1. Drucker, Peter R. Management: Tasks, Responsibilities, Practices. New York: 
 Harper 8c Row, 1973. 
 
 2. Kipling, Rudyard. "The Elephant Child." In Just So Stories for Little Children. 
 Garden City, N.Y.: Doubleday, Doran, 1929, p. 85. "I keep six honest serving-men. They 
 taught me all I knew: Their names are What and Why and When and How and Where and 
 Who." 
 
 3. Council for the Advancement of Small Colleges. Managing Costs and Services in 
 College Libraries: A User's Manual. Washington, D.C.: CASC, 1979. 
 
 4. Markey, Karen. "Analytical Review of Catalog Use Studies" (ERIC, ED 186 041). 
 Columbus, Ohio: OCLC, Research Dept., Office of Planning and Research, 1980. 
 
126 MARY ELLEN JACOB b NEAL K. KASKE 
 
 5. Kaske, Neal K., and Sanders, Nancy P. "On-Line Subject Access: The Human Side of 
 
 the Problem." RQ 20(Fall 1980):52-58; and "Evaluating the Effectiveness of 
 
 Subject Access: The View of the Library Patron." In Communicating Information, vol. 17 
 (Proceedings of the 43rd ASIS Annual Meeting, 5-10 Oct. 1980, Anaheim, Calif.), edited by 
 Alan Benenfield and John E. Kazlauskas, pp. 323-25. White Plains, N.Y.: Knowledge Indus- 
 try Publications, 1980. 
 
 6. Kaske, Neal K., and Ferguson, Douglas. "On-Line Public Access to Library 
 Bibliography Data Bases: Developments, Issues and Priorities" (ERIC ED 195275). Washing- 
 ton, D.C.: Council on Library Resources, 1980. 
 
 7. Kaske, Neal K., and Hildreth, Charles R. "Online Public Access Systems: Data 
 Collection Instruments for Patron and System Evaluation." Dublin, Ohio: OCLC, Office of 
 Research, 1982. 
 
 8. Hildreth, Charles R. Online Public Access Catalogs: The User Interface. Dublin, 
 Ohio: OCLC, 1982. 
 
 9. Kipling, "Elephant Child." 
 
 BIBLIOGRAPHY 
 
 Altman, Ellen O., et al. A Data Gathering and Institutional Manual for Performance 
 
 Measures in Public Libraries. Chicago: Celadon Press, 1976. 
 Association for Systems Management. Management Information Systems. 
 
 Cleveland, Ohio: ASM, 1969. 
 Burns, Robert W., Jr. "Library Use as a Performance Measure: Its Background and 
 
 Rationale." Journal of Academic Librarianship 4(March 1978):4-11. 
 Campbell, H.C. "Methods of Evaluation of Public Library Systems." Public Library 
 
 Quarterly 2(June 1980):35-48. 
 Chacko, George K. Management Information Systems. New York: Petrocelli Books, 
 
 1979. 
 Curran, Charles C., ed. Proceedings of the Library Research Round Table (96th 
 
 Annual Conference of the ALA, Held 17-23 July 1977, at Detroit, Mich.). Ann 
 
 Arbor, Mich.: University Microfilms International, 1979. 
 DeProspo, Ernest R., et al. Performance Measures for Public Libraries. Chicago: 
 
 Public Library Association, 1973. 
 Emerson, Katherin, ed. Proceedings of the Symposium on Measurement of 
 
 Reference. Chicago: Library Administrative Division, ALA, 1974. 
 Golden, Gary A., et al. "Patron Approaches to Serials: A User Study." College & 
 
 Research Libraries 43(Jan. 1982):22-30. 
 Hamburg, Morris, et al. Library Planning and Decision-Making Systems. 
 
 Cambridge, Mass.: MIT Press, 1974. 
 Jacob, M.E.L. A Simple Staffing Model for Predicting Manpower Requirements for 
 
 Library Technical Service Activities. Kinsington, New South Wales: University 
 
 of New South Wales, 1977. 
 Kim, David U. "OCLC-MARC Tapes and Collection Management." Information 
 
 Technology and Libraries 1( March 1982):22-27. 
 Koenig, Michael E. Budgeting Techniques for Libraries and Information Centers. 
 
 New York: Special Libraries Association, 1980. 
 Kriebel, Charles H., et al., eds. Management Information Systems: Progress and 
 
 Perspectives. Pittsburgh, Pa.: Carnegie-Mellon Institute, 1971. 
 Lancaster, F.W. The Measurement and Evaluation of Library Services. Washington, 
 
 D.C.: Information Research Press, 1977. 
 Morse, Phillip M. Library Effectiveness: A Systems Approach. Cambridge, Mass.: 
 
 MIT Press, 1968. 
 Peat, W. Leslie. "The Use of Research Libraries: A Comment about the Pittsburgh 
 
 Study and Its Critics." Journal of Academic Librarianship 7(Sept. 1981):229-31. 
 
MIS: NETWORK ENVIRONMENT 127 
 
 Ross, Joel E., and Murdick, Robert G. An Annotated Bibliography of Management 
 
 Information Systems. Cleveland, Ohio: Association for Systems Management, 
 
 1970. 
 Runyon, Robert S. "Towards the Development of a Library Management 
 
 Information System." College b Research Libraries 42(Nov. 1981):539-48. 
 Wiederkehr, Robert R. Alternatives for Future Library Catalogs: A Cost Model. 
 
 Rockville, Md.: King Research, 1980. 
 Wood, Judith B., et al. "Measurement of Service at a Public Library." Public Library 
 
 Quarterly 2( Summer 1980):49-57. 
 
ROBERT A. KENNEDY 
 
 Director 
 
 Libraries and Information Systems Center 
 Bell Laboratories 
 
 Computer-Derived Management Information 
 in a Special Library 
 
 Introduction 
 
 Not the least of the benefits of automating libraries and information 
 centers is the enhanced ability to monitor processes and services, to collect, 
 structure, analyze, and report critical or useful data hitherto largely 
 unavailable or excessively difficult and costly to obtain. Good manage- 
 ment of information requires good management information 
 information that is as cogent, correct, current, clear, concise, and complete 
 as cost effectiveness and enlightened decision-making demand. Computer- 
 aided information systems offer not only opportunities to gain new 
 insights into the services they support; they challenge the systems designer 
 to build in the feedback necessary to control and improve the systems 
 themselves. 
 
 The focus of this paper is computer-supplied management informa- 
 tion in the special library environment. The particular context is that of an 
 extensively computerized, corporate library network in a large research 
 and development organization Bell Laboratories. 
 
 Library Network Structure and Services 
 
 To help meet the information needs of its 24,000 people (of whom over 
 12,000 are scientists, engineers and managers),* Bell Laboratories has 
 developed a multi-unit library system, structured and managed as a tightly 
 integrated network. The present system is composed of twenty-five librar- 
 
 *A11 data, unless indicated otherwise, are as of April 1982. 
 
 128 
 
SPECIAL LIBRARY APPLICATIONS 129 
 
 ies, a half dozen special information services, and several supporting units 
 serving nineteen Bell Laboratories and Western Electric sites in eight 
 states. 
 
 The primary goal of the library system is to provide to all employees, 
 however distant from company headquarters in Murray Hill, New Jersey, a 
 full spectrum of information services of comparable high quality, at 
 reasonable cost, quickly delivered. To this end, much emphasis is placed 
 on networking, on functional interdependency, resource pooling, respon- 
 sibility sharing, coordinated management, and commonality of systems, 
 standards and goals. To help make networking work, a mix of centralized 
 special services and decentralized standard services is employed, supported 
 by substantial computer, communications, delivery, and management 
 information systems. 
 
 Decentralization is preferred for the basic library functions, i.e., circu- 
 lation and reference services, certain information alerting and online 
 searching services, and the on-site supply of the book, journal and other 
 information services required to handle promptly the majority of local 
 information needs. Collection building, however, is not limited to local 
 needs; each library has responsibilities to the system and, as will be noted 
 later, coordinated resource management is one of the major targets and 
 operational realities of the network. 
 
 Centralized services encompass functions and resources needed 
 throughout the network that would be inefficient or uneconomical or 
 impossible to provide on a local level. Centralization also offers opportu- 
 nities for introducing and insuring common high standards and system- 
 wide service monitoring, as well as supplying the critical mass and 
 economies of scale necessary to justify certain specialized endeavors. 
 Among the operations that are centralized, but not all at one location, are: 
 
 acquisition, classification, cataloging, and the building and mainte- 
 nance of network databases for books, journals, internal technical 
 documents, and other resources; 
 
 information alerting services, including the regular publication of thir- 
 teen major announcement bulletins, a computer-aided system for selec- 
 tively disseminating internal technical documents, and an emerging 
 electronic bulletin system; 
 
 publication of a diversity of specialized information directories, catalogs, 
 indexes, pathfinders, and so forth; 
 
 information retrieval services using highly trained subject specialists, 
 skilled in machine and manual searching, to undertake the more 
 demanding information searches of both bibliographic and numeric 
 databases, compile specific and continuing bibliographies, and 
 supplement reference librarian services; 
 
130 ROBERT A. KENNEDY 
 
 computing information services, including specialized announcement 
 and index services, the handling of external requests for Bell Labs soft- 
 ware, and the operation of libraries specializing in computer informa- 
 tion to support numerous computing activities; 
 
 technical report services furnishing centrally a full range of acquisition, 
 announcement, request processing, and search services on domestic and 
 foreign technical reports of interest to Bell Labs; 
 
 translation services providing oral and written translations by on-site 
 staff or external assignment of all the major languages of science; 
 
 management information services, as outlined in part below, supplying 
 an extensive series of computer-compiled reports on library operations 
 and performance; 
 
 copyright royalty payment operations; and 
 
 information system design and development services charged with the 
 primary responsibility for designing, programming and establishing 
 new and improved information handling systems for the whole network. 
 
 Computer Systems 
 
 Supporting these services and almost every facet of library operations 
 is a complex of computer systems developed by the Bell Labs libraries since 
 the mid-1960s. The principal computer-aided functions and systems are 
 listed in figures 1 and 2 respectively. We shall note only sufficient details 
 about these systems to help place in proper perspective the information 
 management information they have been designed to provide. Additional 
 details about some of the systems and services are contained in an earlier 
 paper and its references. 1 
 
 Management Information 
 
 We shall address, in turn, examples of management information 
 relating to book selection, acquisitions, financial accounting, and catalog- 
 ing; serials control; information alerting, copying and royalty payments; 
 circulation operations, supply services, and resource management; and 
 finally, some retrieval services. 
 
 First, we shall discuss book selection. Two major imperatives apply to 
 collection building in a library network that operates as an integrated 
 system rather than as a loose consortium of essentially separate enterprises: 
 meeting local needs and coordinating total resources in a service- and 
 cost-effective way. Both these requirements have been significantly aided 
 by a computer-supported selection system implemented in Bell Laborato- 
 ries in 1975. In this system, used for both selection and weeding purposes, 
 "selection profiles" have been established for each library. These define 
 
SPECIAL LIBRARY APPLICA TIONS 131 
 
 COLLECTION BUILDING AND UPKEEP 
 
 CATALOGING AND INDEXING 
 
 INFORMATION ALERTING 
 
 DOCUMENT DISSEMINATION 
 
 CIRCULATION CONTROL AND REQUEST PROCESSING 
 
 INFORMATION RETRIEVAL 
 
 PREPARATION OF PUBLICATIONS 
 
 TEXT COMPOSITION AND EDITING 
 
 SERVICE MONITORING AND MANAGEMENT FEEDBACK 
 
 FISCAL CONTROL 
 
 NETWORKING 
 
 COPYRIGHT ROYALTY PAYMENT 
 
 Fig. 1. Computer- Aided Functions in Bell Labs Library Network 
 
 not only the precise subject interest of the location served but the degree of 
 coverage or "collection level" appropriate for that library. Much pertinent 
 management information, including extensive statistical data available 
 from the network's online circulation system on subject usage by technical 
 departments and locations, together with input obtained from users 
 directly and other sources, contribute to the determination of interests. 
 Interests are defined by descriptor and classification number. Collection 
 levels are expressed numerically: = not yet defined; 1 = a representative, 
 basic collection; 2 = research -level coverage; and 3 implies all worthwhile 
 information published worldwide to support long-term research and 
 development interests. For every library, a profile statement and subject 
 index are computer-produced. Figure 3 is a portion of one library's state- 
 ment showing, for example, a level 3 interest in solid state physics. To help 
 coordinate selection on a total network basis, the overall subject interests of 
 all libraries are maintained in a master matrix, available to all, showing 
 the profile level for each library in each subject class. In figure 4, the 
 example, solid state physics (530.4), is shown to be an interest shared on 
 various levels by nine libraries. 
 
732 
 
 ROBERT A. KENNEDY 
 
 BELLTIP Book Acquisition and Cataloging System 
 
 BELLSER Serial Acquisition and Processing System 
 
 BELLREL Loan and Query System 
 
 BELLCAT Catalog Search System 
 
 BELLPAR Publication Announcement System 
 
 BELLTAB Technical Report Announcement System 
 
 MERCURY Internal Document Dissemination System 
 
 BELLPULL Internal Document Request Processing System 
 
 BELLPAY Copyright Royalty Accounting System 
 
 BELLCALL Online Shelflist 
 
 BELDEX Indexing System 
 
 Fig. 2. Computer-Aided Information Systems in Bell Labs Library Network 
 
 Management information systems should, of course, provide periodic 
 checks or analyses to reveal if they are being properly used. The selection 
 profile system has several reports of this kind. One is a "Dollars by Dewey" 
 performance report (see fig. 5) which shows for each subject, the number of 
 items purchased and the dollars spent by each library/level. Total titles 
 held by each library are also given. A companion analysis identifies all 
 purchases made at the zero (undefined) level a situation compelling 
 attention at profile review time. 
 
 Items selected for network libraries under the coordinated profile 
 system are submitted to the BELLTIP book acquisition, accounting and 
 cataloging system operational since January 1972. In this system, input 
 terminals coupled to a large central computer submit order information, 
 receipt and invoice data, cataloging details, file changes, and various 
 queries. Output includes order forms, cancellation and claim notices, 
 financial summaries, "in-process" reports, and much additional data that 
 enable management to monitor purchases, work processes and other areas. 
 Up-to-date process and accounting status reports may be obtained online 
 from any of the many terminals throughout the network. Figure 6 illus- 
 trates use of the online in-process file facility to determine if a particular 
 book has been ordered and, if so, its status. In this example, a keyword 
 
LU 
 
 _J 
 
 LL 
 
 O 
 
 QC 
 O. 
 
 UJ 
 
 > 
 
 UJ 
 
 O 
 
 I- 
 o 
 
 LU 
 LJ 
 
 o 
 
 CO 
 
 * 
 o 
 
 00 * 
 
 co in 
 
 T- co 
 CM in 
 
 to 
 ooo 
 
 in CM 
 r*. ^ 
 
 
 CM 
 
 *- o> ^ r>* 
 ^ co co co 
 
 
 O) 
 
 CM (O CD *- 
 ^ ^ ^ CO 
 
 (O in cq co cq co 
 co co in r^ if> i- 
 
 m co i- o co 
 
 CO CM 
 
 (O(0(ocoin(oininininin!-in(ocominin<0 
 
 a 
 
 s 
 
 QC 
 CD 
 
 oc 
 
 QC 
 
 1 
 
 mmunication) 
 
 
 
 ematical Mode 
 
 
 
 
 
 
 
 
 
 
 v) 
 
 T3 
 
 C 
 3 
 O 
 
 a. 
 
 
 
 
 o 
 
 
 
 ^ 
 
 
 
 > 
 
 
 (/> 
 
 o 
 
 
 
 
 E 
 
 
 SUBJE 
 
 nal Theory (Telec 
 
 lulation Methods 
 
 e (Particles) 
 
 cial Sciences, Mai 
 
 cieties (Scientific! 
 
 lar Batteries 
 
 lid State Chemistr 
 
 lid State Physics 
 
 lid State Reaction 
 
 ubility 
 utions (Electrolyt 
 ving (Problem) 
 jnd 
 
 0) 
 </> 
 
 '5 
 
 z_ 
 
 o 
 
 c 
 
 3 
 
 n 
 
 D) 
 
 C 
 
 co 
 
 <D 
 
 sctra, Infrared 
 
 o 
 O 
 
 I- 
 
 
 
 CO 
 
 "o 
 
 CD 
 
 ectroscopy 
 eech, Analysis 
 
 
 O> 
 
 c. 
 
 N 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o o o o 
 
 O 
 
 a 
 
 a 
 
 a 
 
 Q. O. 
 
 
 CO 
 
 (/) 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 CO CO CO CO 
 
 CO 
 
 CO 
 
 CO 
 
 CO 
 
 CO CO 
 
 
 S 
 
 be 
 
 - 
 
UJ 
 
 ^ 
 
 UJ 
 
 m 
 
 UJ 
 1 
 
 Q 
 
 cc 
 
 J 
 
 {T 
 
 Q. 
 
 k^H 
 
 
 
 - - 
 
 OC 
 
 
 o. 
 
 ^ T- 1- 1- 1- 
 
 1 
 
 2N CMCM CM COCMCMi- 
 
 
 
 z ,- 
 
 1- 
 
 coa 
 
 
 UJ 
 
 J 
 
 
 
 PJ^i-CM CMT-T- i- 
 
 UJ 
 
 <0 
 
 CO 
 
 I 
 
 t^rf 
 
 -"o 
 
 OC 
 
 oc , 
 
 o 
 
 o 
 
 ^ 
 
 X 
 
 ^> 
 
 o 
 
 H 
 
 00 
 
 LU 
 
 
 
 z 
 
 ^ --CM 
 
 OC 
 
 < 
 
 OC 
 
 CD 
 
 _l 
 
 o ^ * 
 
 CM i- 
 
 CM 
 
 ' 
 
 l 
 
 qcoco 
 T-'T^W 
 
 cocococococococococococococococococo 
 
 CD 
 
CO 
 CD 
 
 CM 
 
 LU 
 
 O 
 
 z 
 
 ^ 
 
 
 
 h-CMOCOh-mCOCM 
 
 or^ocMCMomcM 
 
 
 
 
 LE PERFORN 
 
 BY DEWEY) 
 
 0) 
 
 E 
 
 0) 
 
 NOOCO^OCO^ 1 
 (O O CM C3) ^ CO ^ 
 T- CM CO 
 
 h-G)O^CO T "' T ~CM 
 CM 
 
 c 
 o 
 
 '< 
 
 
 
 c 
 o 
 
 '^ 
 
 o 
 
 ^^ 
 
 
 
 
 ^ 
 
 c 
 
 ^" 
 
 LL 
 
 (/) 
 
 
 
 o 
 
 o 
 
 o 
 
 o 
 
 DC 
 
 
 
 o 
 
 5 
 
 ^ 
 
 DC 
 
 < 
 
 > 
 
 (/) 
 o 
 
 Q> 
 
 ^> 
 
 0) 
 
 CL 
 
 -J 
 
 CD 
 
 
 ^ 
 
 "o 
 
 'to 
 
 
 j 
 
 -J 
 
 cMCMi-i-eoi-i-*- 5. 
 
 CD 
 
 o 
 
 c 
 
 o 
 
 o 
 o^ 
 
 1 
 
 CD 
 
 1 
 -IOIZI>CCQ ^ 
 <X~~52 Q - CC 2 
 
 c 
 
 Q) 
 
 o 
 
 
 
 Q. 
 
 o 
 
 
 k_ 
 
 n 
 "T 
 
 CD 
 C/) 
 
 Q. 
 0) 
 
 0) 
 (A 
 
 E 
 
 
 
 UJ 
 
 1 
 
 
 ^j 
 
 ^3 
 
 cc 
 
 cc 
 
 o 
 
 J 
 UJ 
 
 
 
 "o 
 
 - 
 
 CM 
 
 CO 
 
 
 
 
 (/) 
 
 ^) 
 
 TJ 
 
 ^> 
 
 Mi 
 
 
 O 
 
 T- ^ 
 
 > 
 
 > 
 
 > 
 
 
 
 Z 
 
 ^ ^t 
 
 0) 
 
 I 
 
 0) 
 
 1 
 
 CD 
 
 1 
 
 
 
 o 
 
 6 o 
 
 J 
 
 J 
 
 _J 
 
 
 
 Q 
 
 CO CO 
 
 
 
 
 
 
 Q 
 
 m in 
 
 
 
 
 c 
 d 
 
 u 
 
136 ROBERT A. KENNEDY 
 
 search was made for the book, Soul of a New Machine; two libraries are 
 shown to have ordered and just received the item. Another online report is 
 particularly useful in monitoring the overall status of items flowing 
 through the acquisitions/invoicing/cataloging system of BELLTIP. This 
 report, the In Process File Scoreboard, permits a manager not only to 
 determine how many items are in any given status (e.g., on order, in 
 cataloging, etc.) but to compare the current status with a selected earlier 
 date, say one month ago, as shown in figure 7. Scoreboard reports not only 
 book copies, or book titles as appropriate, but machine words i.e., disk 
 space in use for the system manager. 
 
 Other insight into problem situations in the BELLTIP world e.g., 
 books remaining too long in a particular state such as cataloging or 
 invoicing is provided automatically in a series of specific offline reports. 
 On the financial side, online accounting reports, not shown, give to-the- 
 moment information on payments, commitments, budgets, and percent 
 committed for any or all libraries specified. Supplier account information 
 is similarly available online. Still another report, available in both online 
 and offline versions, provides valuable information on vendor perfor- 
 mance. This report summarizes average order costs, discount percentages, 
 delivery times, and number of claims and cancellations for each vendor 
 over a period of time; the document's usefulness to management in renew- 
 ing purchase orders and maintaining effective supply sources is obvious 
 and substantial. 
 
 Several of the information tools helpful to catalogers in managing 
 their activities may be of interest. One is a simple online facility, 
 BELLCALL, for determining if a newly-selected call number is already in 
 use; the video display response identifies the two closest neighbors on 
 either side (see fig. 8). Another tool, useful in managing the large, special- 
 ized subject headings authority database, is a periodic printout of all the. 
 headings used with a given class number (see fig. 9) and another printout of 
 all the class numbers associated with a given heading. These instruments 
 are helpful in developing standardized headings for the selection profile 
 system, for reclassifying parts of the collection, and for adapting or refin- 
 ing subject headings to the specificity required in a high-technology 
 environment. Still another management report useful to catalogers, but 
 especially to library supervisors for collection management, is the Related 
 Editions Report which identifies all the editions held of titles held in more 
 than one edition in any library in the network; multiple editions may or 
 may not be justified, but they should be scrutinized. This report, and 
 another listing books by publication date, are of substantial help in keep- 
 ing collection growth, and shelf space, under control. The principal 
 management aids generated by the BELLTIP acquisitions, cataloging and 
 accounting system are summarized in figure 10. 
 

 rH 
 
 rH 
 
 a. 
 o 
 
 E- 
 
 w 
 
 0) 
 
 >, 
 
 4J 
 
 
 in 
 
 rH 
 
 1 
 vc 
 
 rH 
 
 m 
 
 Z 
 
 CQ 
 W 
 
 rH 
 (/) 
 
 
 
 
 VO 
 
 1 
 
 00 
 
 z 
 u 
 u 
 
 o 
 
 D. A.D. PO/VEN INV PRICE B-NUMBER SUB 
 63 07 P 12.55 199390D 01 
 77 32 P 10.46 199390D 03 
 
 
 
 M 
 
 1 
 
 CM 
 ^C 
 
 OJ 
 
 
 Cu 
 
 
 
 >v^ 
 
 
 
 
 
 51 
 
 51 
 
 
 
 
 w 
 
 a, 
 
 
 0% 
 i 1 
 
 
 
 
 
 Cfl CM 
 
 CM 
 
 
 
 
 W 
 
 o 
 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 a> 
 
 
 
 ^T 
 
 
 
 
 
 EH 
 
 
 
 
 o 
 
 w 
 
 
 VO 
 
 
 
 
 
 
 < W 
 
 CO 
 
 bb 
 
 
 o 
 
 
 0) 
 
 
 
 0) 
 
 
 
 
 EH 
 
 
 IS 
 
 
 V_i 
 
 o 
 
 C 
 
 rH 
 
 c 
 
 
 
 
 CO 
 
 
 
 
 CU 
 
 E- 
 
 rH 
 
 51 
 
 rH 
 
 
 
 
 
 
 
 CM 
 
 
 
 f^ 
 
 C. 
 
 r; 
 
 
 
 
 
 
 
 
 00 
 
 C 
 
 
 O 
 
 
 
 
 
 
 Q 
 
 . m 
 
 ro 
 
 
 
 rH 
 
 . 
 
 
 .. 
 
 (D 
 
 
 
 CO 
 
 Q ^r 
 
 VD 
 
 
 rH 
 
 
 _ 
 
 
 
 O 
 
 E 
 
 
 
 
 5) 
 
 51 
 
 
 
 OJ 
 
 CU 
 
 
 Z 
 
 
 
 
 
 O CM 
 
 CM 
 
 
 * 
 
 c 
 
 _] 
 
 Z r-t 
 
 
 u 
 
 
 
 
 
 
 
 5) 
 
 rH 
 
 CO 
 
 M 3 
 
 J 
 
 D 
 
 
 
 
 
 
 
 ro 
 
 rH 
 
 X i-l 
 
 X O 
 
 J 
 
 C 
 
 
 
 
 
 CO 
 
 
 
 
 C 
 
 3 
 
 u w 
 
 <c 
 
 
 
 
 
 a 
 
 cu 
 
 
 
 jj 
 
 o 
 
 
 
 *^ "\ 
 
 u 
 
 (0 
 
 E- 
 
 
 CO 
 
 >4 
 
 a: 
 
 
 fO 
 
 
 * ""O QJ CO 
 
 Z. X 
 
 
 
 
 
 CJ> 
 
 E- 
 
 
 
 C 
 
 a> 
 
 (DO) C 
 
 
 
 UH 
 
 . 
 
 
 
 (N 
 
 
 
 
 
 
 O, <U -rH 
 
 
 o 
 
 
 
 i_i 
 
 OJ 
 
 
 
 
 
 0) 
 
 4J 
 
 >> u -c 
 
 
 51 
 
 
 QJ 
 
 rH 
 
 
 
 
 
 
 U-i 3 
 
 
 4J O O r- 1 
 
 in rH 
 
 C"> 
 
 rH 
 
 T3 
 
 4J 
 
 rH 
 
 
 
 
 
 a EH 
 
 o 
 
 V_i (Q 
 
 in 
 
 m 
 
 3 
 
 T3 
 
 4- 
 
 00 
 
 3 
 
 
 
 rH 
 
 
 a e 
 
 
 ^ 
 
 
 
 rH 
 
 r-( 
 
 O^ 
 
 O 
 
 Q 
 
 
 ^S *""^ 
 
 
 a 
 
 
 CTv 
 
 
 tii 
 
 J 
 
 rH 
 
 CO 
 
 M 
 
 
 <D m 
 
 a> 
 
 rH >l rH 
 
 
 rH 
 
 
 
 
 
 ce 
 
 E- 
 
 
 C 
 
 E 
 
 o nj 3 
 
 
 
 
 
 
 
 
 
 
 
 rH rH 
 
 O 
 
 E 
 
 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 rH 51 
 
 O 
 
 U) rH 
 
 CM m 
 
 O 
 
 j 
 
 X 
 
 J 
 
 a: 
 
 rH 
 
 rH 
 
 
 C 
 
 II 
 
 V- 3 \ 
 
 
 z 
 
 E- 
 
 E- 
 
 CQ 
 
 <i 
 
 2Q 51 
 
 51 
 
 
 O O^ 
 
 a> 
 
 O O -* 
 
 
 1 
 
 ii 
 
 HD 
 
 r^ 
 
 CO 
 
 rH CC 
 
 CQ 
 
 
 * Q 
 
 
 b X II 
 
 
 CQ 
 
 E- 
 
 < 
 
 cu 
 
 >< 
 
 J Q 
 
 
 
 
138 
 
 ROBERT A. KENNEDY 
 
 Please give cutoff date (today is 3728) 
 
 = 3698 
 
 The following copies are inadequate for order 
 
 326025A 1 
 
 326600A 1 
 
 The following titles are inadequate for cataloging 
 
 199297C 
 
 The In Process File includes: 
 
 Copies 
 
 Requisitioned 
 
 On Order 
 
 Claimed 
 
 Standing Orders 
 
 In Cataloging 
 
 In Preparations 
 
 Sent 
 
 Sent Direct 
 
 Cancelled 
 
 Aborted 
 Titles 
 
 Supporting Order 
 
 Orig. Cataloging 
 
 Catalog Changes 
 Invoice Traces 
 
 Total Now 
 # entries # words 
 
 136 
 3944 
 
 665 
 51 
 
 1193 
 1179 
 1842 
 
 104 
 
 121 
 
 
 6208 
 2524 
 1271 
 
 234 
 
 3768 
 
 69 
 
 82550 
 
 2181 
 
 17864 
 
 476 
 
 1041 
 
 50 
 
 34095 
 
 367 
 
 35634 
 
 35 
 
 55278 
 
 4 
 
 3011 
 
 1 
 
 2823 
 
 
 
 
 
 
 
 345104 
 
 166045 
 
 27444 
 
 3723 
 
 Prior to Day 3698 
 # entries # words 
 
 1608 
 
 46012 
 
 13356 
 
 1021 
 
 10255 
 
 1082 
 
 124 
 
 31 
 
 
 
 
 
 260004 
 
 140259 
 
 
 
 4547 
 
 2083 
 
 
 
 Fig. 7. Online In Process File Scoreboard 
 
 Comparable to the BELLTIP system for books is the BELLSER 
 system for serials acquisition, cataloging and financial control. One of its 
 management facilities is called MONEY. This provides online displays of 
 library accounts, vendor accounts and other serials data, including antici- 
 pated inflation factors for budgeting purposes (see fig. 11). The cost of 
 serials in libraries supporting advanced technology explorations tend to be 
 significant and demand close management attention to keep them reason- 
 able. The cost problem is compounded in a large multi-library environ- 
 ment. One practice followed in the Bell Labs Library Network is to have a 
 formal, face-to-face annual meeting of librarians to review serial orders 
 and negotiate "adds and drops" in a coordinated manner. To aid this 
 process, the decision-makers have much management information, 
 
SPECIAL LIBRARY APPLICATIONS 139 
 
 = 001.6424P27/F79p 
 
 Nothing on file for call number #001.6424P27/F79p 
 
 Books with close call numbers follow 
 
 001.6424P27/E36p + 153174 
 
 EISENBACH, S./ PASCAL PROGRAM 
 
 001.6424P27/F49p + 122510 
 
 FINDLAY, W./ PASCAL 
 
 001.6424P27/G74I + 176818 
 
 GRAHAM, N. INTROD PASCAL 
 
 001.6424P27/G87p + 149623 
 
 GROGONO, P. PROGRAMMING IN PASCAL 
 
 Fig. 8. BELLCALL Online Call Number Facility 
 
 including circulation and alerting bulletin statistics, user reactions, etc. 
 Especially helpful are two BELLSER reports: one an alpha listing of all 
 current serial subscriptions and their prices; the other, a price-ordered list, 
 high to low, of all serials (see fig. 12). Clearly, the high-cost journals at the 
 top of the list get particular scrutiny. 
 
 Strong emphasis is placed in the Bell Labs Library Network on 
 information alerting and dissemination services. As already noted, thirteen 
 different computer-compiled announcement bulletins addressed to partic- 
 ular subjects and audiences are regularly produced. Supplementing these 
 network media are a number of local library bulletins that focus on the 
 special interests of a given laboratory location. Another major component 
 of the current information alerting service is MERCURY, a computer- 
 driven system for selectively disseminating internal reports. Electronic 
 versions of the bulletins are also being developed. 
 
 How are these alerting services kept on target? Substantial manage- 
 ment information is available to the editors, primarily from the computer 
 
equency 
 
 wave 
 equency 
 
 
 equency 
 
 si on 
 
 
 e 
 
 
 C 
 
 
 -. 
 
 
 > 
 
 CO fi (0 
 
 y o u 
 
 O C "4 -H 3> U 
 
 ^ O C <J 2 "4 
 
 n -H o ^> -H > 
 
 ITJ 4J U U U J 
 
 ulation 
 
 US D 
 
 w JJ w 
 
 g 
 
 o 
 
 H 
 
 2 
 
 4J U 
 
 C 
 
 -0 O 
 U -H 
 
 
 u 
 
 Ll ffl 4J H ^3 0* 
 
 o> y c a c 
 
 3 
 
 H U 
 
 1) 
 
 
 
 
 4J 
 
 DJ CT* 
 
 o 
 
 
 
 C 1) 
 
 o ITJ o> 3 q M D -H 
 
 s 
 
 -3 W .C 
 
 % 3 
 
 w s. 
 
 7) 
 
 fl 
 
 i/ U 
 
 u - 
 
 C TJ 
 
 
 3 
 
 H ^ui-l -H (II *J *J U) 
 
 M 
 
 CT '!> -C T> 
 
 3 a- 
 
 Vt C^ 
 
 c 
 
 U 
 
 y -H 
 
 > *J 
 
 3 W 
 
 to : 
 
 N-3 3 1 
 
 uoie^'>'BOtnai SN -H 
 
 >, 
 
 H .* CO H 
 
 T i) 
 
 H -^ M 
 
 LJ _ 
 
 
 9 1! U 
 i LJ .1 
 
 --n 
 
 1 L, 
 
 c to < 
 / ^ 
 
 J 4J V 
 
 - !< Ll 
 
 a)kJ _3'* ; ''^" '' :!U4J 
 
 
 I >-*Mil^4j-H.iH(09-HUtfl | 4iaiQQi4l 
 
 i s U 
 
 to (0 
 
 5 
 
 M E -H 
 
 0)(03>*JOOOi3>flCT 1 33 
 to O > C "< H - -H 0) O 5* ^"O 
 3C(T3T5'*3"O'C3^H'HL|>OO 
 
 OO99idOV^4viE 
 
 a 
 
 X 
 
 C CO N 
 
 O L, U 
 
 H 0> Ll 
 
 re 
 U 
 
 1 
 
 D 
 
 y a 4J 
 
 y a. rH 
 
 u u to e S 
 
 ' >t TJ -a 0) M w 
 
 4J Oi Oj I -H S C CO 
 
 N ^J 
 
 J M H 
 
 LI y 
 
 v - 
 
 2 u 
 
 - - _ 
 
 o -a y 
 
 S jJ M S 
 
 - ----- ^ 
 
 TJ -i 13 'H Ll y T3 
 
 W > -u fc-i > * 
 
 a 3 a $ -*10 S * 
 
 LI 3 
 
 o u 
 
 10 
 1) U 
 
 N 
 
 ~^ ~ '~ 
 
 U C Q 
 c < 
 
 o .Si, 
 
 s^ 
 
 << 
 
Selection Profile Management - Indexes, Matrices, Dollars 
 
 Online In Process File - Order Status, etc. 
 
 Online Accounting - Library, Budgets, Expenditures 
 
 Online Process Overview (Scoreboard) 
 
 Monitoring/Action Reports - Ordering, Cataloging, Invoicing, etc. 
 
 Vendor Performance Analysis 
 
 Call Number Usage (BELLCALL) 
 
 Subject Heading/Classification Number Analyses 
 
 Related Editions, Dated Publications 
 
 Fig. 10. Some Management Aids from the BELLTIP Acquisitions, 
 Cataloging, Accounting System 
 
 Displays library accounts for serials 
 
 Expenditures, encumbrances, budgets, % committed 
 
 For all libraries, some, or single library 
 
 Displays vendor accounts 
 
 For all or specified vendors 
 
 For all or specified libraries 
 
 Displays vendor inflation factor 
 
 For all or specified vendors 
 
 Handles online changes in cost and budget data 
 
 Fig. 11. MONEY, an Online Management Information Facility in the 
 BELLSER Serials System 
 
CIB 
 
 1 
 
 MC 
 
 $ 
 
 HL 
 
 1 
 
 MC 
 
 $ 
 
 IH 
 
 1 
 
 F 
 
 $ 
 
 IW 
 
 1 
 
 F 
 
 $ 
 
 PY 
 
 1 
 
 MC 
 
 $ 
 
 244784v Journal of Plasma Physics 
 
 MH 1 MC $ 354.00 
 278640v Theoretical Computer Science 
 
 350.00 
 350.00 
 350.00 
 350.00 
 350.00 
 210112h Biochemical and Biophysical Research 
 
 Communications 
 
 MH 1 MC $ 345.00 
 237352n Soviet Materials Science 
 
 MH 1 MC $ 345.00 
 220770s Dissertation Abstracts International. 
 
 C. (European Abstracts) . 
 
 HO 1 MC $ 345.00 
 
 MR 1 MC $ 345.00 
 278656m Theoretical and Experimental Chemistry 
 
 MH 1 MC $ 345.00 
 219584d Cybernetics 
 
 LS 1 MC $ 345.00 
 
 MH 1 MC $ 345.00 
 215600y Chemical Engineering Science 
 
 MH 1 E $ 340.00 
 
 PR 1 E $ 340.00 
 218896f CRC Critical Reviews in Analytical 
 
 Chemistry 
 
 MH 1 F $ 332.00 
 200015h CRC Critical Reviews in Biochemistry 
 
 MH 1 F $ 332.00 
 279680a Transactions of the American 
 
 Mathematical Society 
 
 HO 1 MC $ 330.00 
 
 MH 1 MC $ 330.00 
 
 WH 1 MC $ 330.00 
 
 Fig. 12. Price-Ordered Subscription List 
 
SPECIAL LIBRAR Y APPLICA TIONS 143 
 
 systems used to process the article request traffic and to record copyright 
 royalty payment obligations. The bulletin request program, for example, 
 employs information about the journal and the requester (e.g., depart- 
 ment, location) to compile a rich statistical record of usage; among the 
 analyses routinely made are various rank orderings of the journals covered 
 in the bulletins by articles announced, total requests, announced/re- 
 quest ratios, etc. These data greatly assist in determining which of the 
 many competing journals should be regularly indexed; although heavy 
 use is made of external data sources, many journals must still be indexed 
 in-house and careful attention to profitability is necessary. A similar 
 approach is taken for the technical reports bulletin; analyses of the "hit" 
 rates of each of the bulletin's many subject categories help in refining 
 coverage and maintaining an overall hit rate of 80 percent or better. 
 
 BELLPAY, the copyright royalty payment system begun in 1978, 
 provides further valuable insights into journal usage. One report from the 
 system (see fig. 13) summarizes the demand for journal articles by time 
 i.e., by the date of publication. This provides a perspective of substantial 
 consequence in managing costly serial resources. 
 
 We now consider the management information which the online 
 circulation system (BELLREL) has been designed to supply. The princi- 
 pal reports are listed in figure 14. The most important is the Titles in 
 Demand list, produced weekly to identify all book titles for which there is a 
 total of five or more people waiting anywhere in the network (see. fig. 15). 
 The two-line entry for each title shows the number of people waiting at 
 each location, the number of copies available, copies on order (as learned 
 from the BELLTIP acquisition system) or missing, and the ratio of copies 
 to requesters in the waiting queue. (In figure 15, the book by Ball, Algo- 
 rithms for RPN Calculators, has forty-six people waiting, sixcopiesavail- 
 able or on order, and hence a ratio of eight.) Using this list, library 
 supervisors hold a weekly teleconference to decide what additional pur- 
 chases should be made to help meet demands and keep response time short. 
 Counter balancing this "hot" list is a "cold" list, the Zero Activity Report, 
 run at least annually to identify titles that have had little or no recorded use 
 in a specified time span. This list, along with such other management data 
 as the principal users of a given subject class, are distinctly helpful in 
 purging the collection, conserving space and saving money. 
 
 Several of the other BELLREL management reports listed in figure 1 4 
 might be noted. The weekly Reserve Queue Aging Report identifies 
 requests not satisfied within a specified time period because of some 
 hangup such as the loss of the book. The Get Off the Shelf Report points to 
 all cases where the system says (1) a request exists, (2) a copy is, in fact, 
 available to meet the need, but (3) for some reason the item has not yet been 
 
144 
 
 ROBERT A. KENNEDY 
 
 230320 IEEE TRANS COMPUT 
 
 CURR 1 Yr. 3 Yr. 6 Yr. 
 
 10 Yr. 20 Yr. >20Yr. TOTAL 
 
 AK 
 
 
 
 
 
 1 
 
 
 
 1 
 
 
 
 
 
 2 
 
 AL 
 
 46 
 
 10 
 
 3 
 
 1 
 
 4 
 
 4 
 
 
 
 67 
 
 CB 
 
 18 
 
 11 
 
 
 
 
 
 1 
 
 
 
 
 
 30 
 
 OR 
 
 38 
 
 
 
 1 
 
 1 
 
 1 
 
 
 
 
 
 50 
 
 EL 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 FJ 
 
 54 
 
 29 
 
 
 
 
 
 
 
 
 
 
 
 83 
 
 HL 
 
 3 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 6 
 
 HO 
 
 153 
 
 95 
 
 16 
 
 5 
 
 12 
 
 3 
 
 
 
 283 
 
 HP 
 
 20 
 
 17 
 
 7 
 
 8 
 
 8 
 
 1 
 
 
 
 61 
 
 IH 
 
 64 
 
 60 
 
 2 
 
 
 
 2 
 
 
 
 
 
 108 
 
 IN 
 
 16 
 
 34 
 
 
 
 
 
 
 
 
 
 
 
 49 
 
 IW 
 
 66 
 
 43 
 
 
 
 
 
 
 
 
 
 
 
 98 
 
 MH 
 
 166 
 
 98 
 
 14 
 
 10 
 
 26 
 
 4 
 
 
 
 318 
 
 MV 
 
 10 
 
 
 
 
 
 
 
 
 
 
 
 
 
 10 
 
 NP 
 
 8 
 
 4 
 
 
 
 
 
 
 
 
 
 
 
 12 
 
 PR 
 
 3 
 
 27 
 
 1 
 
 1 
 
 1 
 
 
 
 
 
 33 
 
 PY 
 
 27 
 
 26 
 
 
 
 4 
 
 3 
 
 
 
 
 
 60 
 
 RD 
 
 12 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 13 
 
 8T 
 
 2 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2 
 
 WB 
 
 7 
 
 7 
 
 
 
 
 
 2 
 
 
 
 
 
 16 
 
 WH 
 
 48 
 
 19 
 
 1 
 
 6 
 
 9 
 
 2 
 
 
 
 85 
 
 XX 
 
 14 
 
 22 
 
 
 
 
 
 
 
 
 
 
 
 36 
 
 763 
 
 606 
 
 45 
 
 36 
 
 70 
 
 14 
 
 Fig. 13. Journal Usage Data from the BELLPAY Copyright 
 Royalty Payment System 
 
 1423 
 
 supplied. Missing Item Reports, produced every two weeks in class number 
 order, are used routinely to search shelves. In addition to these BELLREL 
 reports, much other circulation information is available in a variety of 
 correlations of subject, library, time, and borrower. One of these is the 
 Loan History Report, a subject-classed listing of the materials borrowed by 
 a specified technical department. This report is particularly helpful in 
 establishing an appropriate onsite collection when a technical department 
 moves to another laboratory location. 
 
 Information retrieval services at Bell Laboratories both supply and 
 apply management information to good effect. Management information 
 
SPECIAL LIBRARY APPLICATIONS 145 
 
 Titles in Demand 
 
 Zero Activity Report 
 
 Reserve Queue Aging 
 
 Get Off the Shelf 
 
 Missing Items 
 
 Circulation Data - Library, Subject, User Dept. Correlations 
 
 Fig. 14. Some Management Aids from the BELLREL Circulation System 
 
 is, of course, often the subject of online searches. In addition to biblio- 
 graphic retrieval, however, bibliometric studies are frequently a most 
 appropriate exercise of interactive searching powers. Quantitative anal- 
 yses of a body of literature to assemble, say, data on the leading crystallo- 
 graphic journals or the principal contributors to the field of magnetic 
 bubbles, can be of much interest to information, and other managers. 
 Online searching also generates statistics helpful in monitoring and meas- 
 uring the systems and databases used, cost trends, searcher activity by 
 library and time and type of search, and so on. Extensive compilations of 
 these kinds of data are regularly given to management by the Bell Labs 
 Information Retrieval Service. 
 
 Conclusion 
 
 We have touched briefly on some of the numerous variants of manage- 
 ment information reports that can be made available in a computer- 
 supported special library. Much of the satisfaction of working with 
 automated systems derives from power and responsiveness of the systems in 
 providing decision-aiding data. As most information managers recognize, 
 however, the very ease with which elegant compilations of all kinds can be 
 
2 * 
 
 s 
 
 i 1 1 
 
 !R 8 
 
 LU o 
 
 T * 
 
 a 
 
 <u 
 
 
 
 T3 
 
 ~ 
 
 Q 
 
 c 
 
 
 
 Sw 3 -x^ ^ ^T " ^ Ok ^" *^ 
 
 -S S ^S ^S 5S ^3 
 
 > co S o " cb o e 5* 
 
 NO Nn NM ^ n 00 
 
SPECIAL LIBRARY APPLICATIONS 147 
 
 produced may encourage excesses in reporting. The results may be 
 unhappy (see. fig. 16). 
 
 Excesses in Information Supply, Frequency, Complexity 
 
 I 
 Information Overload, Noise 
 
 1 
 Information Disregard, Rejection 
 
 1 
 Decisions On Assumptions, Past History, Hopes, Wind Direction 
 
 i 
 Poor Management, Trouble 
 
 Fig. 16. Management Information Systems Perils 
 
 Professionals engaged in the uncertainties of the information transfer 
 process, striving to couple information need to information source pre- 
 cisely, swiftly and economically, need all the help they can get. The value 
 to information managers of a carefully defined, coherent management 
 information program can hardly be overstated. 
 
 REFERENCE 
 
 1. Kennedy, Robert A. "Bell Laboratories Library Network." In The Special Library 
 Role in Networks, edited by Robert W. Gibson, Jr., pp. 17-36. New York: Special Libraries 
 Association, 1980. 
 
JANE L. CALDWELL 
 
 Coordinator of Professional Activities 
 Bibliographic Retrieval Services 
 
 Management Information from 
 Bibliographic Databases 
 
 This year's clinic dealing with the use of information for library manage- 
 ment is very exciting. Perhaps the most exciting element is the fact that we 
 are not talking in futuristic terms. The advent of the personal computer 
 means that almost any library can afford an in-house database as well as 
 access to outside databases. Sophisticated software, such as 
 BRS/SEARCH, will soon be available for use on one's local mini- or 
 microcomputer. This promises local access to the online catalog, an auto- 
 mated circulation control system, serials check-in, and the ability to store 
 and manipulate library statistics and other types of management informa- 
 tion. Libraries will have much better control of their own statistics and 
 have surer methods to determine collection usage, user satisfaction, 
 employee performance, etc. 
 
 Today, I am going to take a somewhat different, I hope, approach to 
 the subject of management information than has been taken thus far, and 
 describe information that is already available, easily manipulated and 
 frequently overlooked. This information is available to all of those whose 
 libraries are now accessing online databases through one or more of the 
 vendors of online databases. Some of those vendors include BRS, Dialog, 
 NLM, and SDC. There is a wealth of information available to the user of 
 online databases that they may have overlooked. 
 
 First of all, what is a database? The simple definition is that it is a 
 collection of data in machine-readable form. This data could be available 
 in a batch mode or an online mode. I will limit my discussion today to 
 online databases that are available on a dial-up basis from one of the search 
 services I have mentioned. When I say that a database is available online I 
 mean that it is possible to query the database, get results immediately and 
 
 148 
 
BIBLIOGRAPHIC DATABASES 149 
 
 interact with that information all from a local terminal. An online data- 
 base from one of the search services could have a hard-copy counterpart 
 such as the ERIC database that corresponds to the print indexes of Re- 
 sources in Education and the Current Index to Journals in Education. 
 Sometimes there is no exact print counterpart as in the case of the Dow 
 Jones database; sometimes, there is no print counterpart at all, as in the 
 case of the Pre-Med databases. A database may be bibliographic, meaning 
 that it includes citations, often with abstracts or summaries, to printed 
 journals, monographs, technical reports, patents, etc. A database may be 
 nonbibliographic, meaning that it contains data, usually numeric or 
 directory-type information. A database may be full-text, meaning that the 
 full journal article or newspaper story is available online. Examples would 
 be the American Chemical Society's project to put their sixteen chemical 
 journals online. Another would be the full text of the Harvard Business 
 Review that will be available online from BRS. In libraries today, all types 
 of databases are being searched to produce comprehensive bibliographies 
 or to answer specific reference questions. 
 
 Online database searching began, at least in an experimental mode, in 
 the early 1960s. In the late sixties some government-produced files, notably 
 that of NASA, were the first to be made available in an "operational" 
 mode. In the early seventies, commercial search services offered access 
 through telecommunications systems to, at first, only a few databases. The 
 number of databases has rapidly expanded and from one's terminal one 
 now has access to hundreds of different databases. When online databases 
 first became commercially available, the softwr re was not nearly as flexible 
 as it is now, response time as short as it is now, and the costs for an online 
 search as low as they are today. When online searching first became 
 popular, databases were used primarily to produce comprehensive bibliog- 
 raphies on specific topics. Later, as online search analysts became more 
 proficient and searching became an accepted research tool, online data- 
 bases were regularly accessed to answer the "quick" or "ready" reference 
 questions. 
 
 Why use an online database to answer a reference question that 
 perhaps could be answered from a directory or encyclopedia? The answer 
 lies in the fact that with an online database, all the information from the 
 record some information that may never go into the printed version is 
 directly searchable. (There may be variations between the vendors of online 
 services it depends upon the way that the database has been structured by 
 the vendor and what information has gone into the dictionary or inverted 
 file, or basic index.) Information is available in an online database that is 
 not available in a printed index (e.g., titles, journal citation, publisher's 
 name, price, author's affiliation, funding agencies, etc.). Often this infor- 
 mation could be accessible through a printed source, but it would be time 
 
150 JANE L. CALDWELL 
 
 prohibitive to find this information. Besides the obvious advantage of 
 Boolean operators OR, AND, NOT or pull together synonyms or com- 
 bine and remove unwanted concepts positional or proximity operators 
 allow the searcher to be more specific or to search for concepts that are too 
 new to have standardized terminology. 
 
 So far, we are talking about what I consider to be the traditional 
 approach to the use of online databases. Databases are primarily used to 
 produce comprehensive bibliographies with the costs often passed back to 
 the user. Databases are frequently used for reference purposes, often with 
 the charges absorbed by the reference department. Usually money is allo- 
 cated solely for this purpose and it is up to the reference department to 
 determine if accessing an online database is the appropriate way to answer 
 a specific question. 
 
 Not long after the introduction of online database searching, many 
 librarians, and researchers in other fields as well, discovered that there were 
 ways that online databases, and the statistics they generated, could be 
 approached in less traditional ways to provide all types of information. 
 Donald Hawkins wrote a paper in 1977, published in JASIS, describing 
 some of these unconventional uses of databases. The title of his paper was 
 "Unconventional Uses of On-line Information Retrieval Systems: On-line 
 Bibliometric Studies." 1 In it, he described ways to use databases to quanti- 
 tatively analyze the bibliometric features of the literature. By using online 
 databases we can explore, at no great expense in terms of time or money, 
 aspects of the literature in a specific discipline. Some of the characteristics 
 of the literature that we can explore that will have great impact on one's life 
 as a library decision-maker include information of an evaluative nature 
 particularly with applications for collection development. Other informa- 
 tion is available such as information useful for planning future use of the 
 library collection, information to plan and budget for use of library ser- 
 vices, and information to use in predicting trends in book and journal 
 publishing. Information from online databases can also be used to evalu- 
 ate personnel both library personnel and staff outside the department. 
 
 The obvious use of databases for library management is using data- 
 bases with the subject emphasis on management, such as Management 
 Contents or AEll INFORM. 
 
 In my talk today, I want to take a very practical approach and describe 
 ways that one can use online databases to provide information that has 
 great implications for library management and for management of one's 
 institution. I am not going to talk about the obvious using databases 
 with the subject emphasis on management, such as AEll INFORM or 
 Management Contents, that provide references to printed literature on 
 management topics. Rather, I am going to discuss the use of databases to 
 quantitatively explore aspects of the literature that have implications for 
 
BIBLIOGRAPHIC DATABASES 
 
 collection development, for determining how clientele use the services, and 
 how their use of the literature will impact one's library. I am also going to 
 describe how the statistics generated by online database searching can be 
 effectively used for planning, budgeting and evaluating one's library ser- 
 vices. The information I am going to discuss today is available from other 
 databases and other search services, but I will be speaking primarily of 
 access to BRS databases. I am, of course, more familiar with BRS databases 
 and slightly prejudiced toward BRS as an innovative search service. How- 
 ever, I do believe that BRS has been the most responsive to the needs of 
 managers in developing extra services such as our collection devlopment 
 service, faculty bibliography service, electronic newsletter service 
 (BRS/ Alert), and our online accounts file. 
 
 Getting back to the primary subject, I would like to show how online 
 databases can be used for different evaluative or bibliometric studies. 
 Online databases can be used for citation analysis which could have 
 valuable applications in collection development. 
 
 Bibliometrics has existed as a field of study since 1917. Bibliometrics 
 became more sophisticated when Bradford, in the 1930s, established a basic 
 law of bibliometrics Bradford's Law of Scattering. Bradford discovered 
 that, by arranging journals in descending order of productivity of articles 
 on some subject, it is possible to identify a "core" of journals that publish 
 most of the articles on a particular subject. Trueswell took this one step 
 further and suggested an 80/20 rule 20 percent of the journals in a given 
 field account for 80 percent of the relevant literature. The manual work 
 that can go into identifying some core list is beyond the resources of most 
 library managers. Online systems can provide a method to simplify the 
 identification of this core. Here is a method developed in part by Donald 
 Hawkins, Jane Caldwell and Celia Ellingson: 
 
 1. Create a set of all journal articles on a topic. This would be a broad 
 search to account for variations in terminology. Use paragraph 
 qualification to make sure the articles themselves will deal with the 
 subject matter rather than having the terms appear in the journal titles. 
 
 2. Sort by journal titles (if this feature is available on your search 
 system it is on BRS) and print journal citations. 
 
 3. Run the same search on all relevant databases. 
 
 4. Rank order journals by number of articles on topic. 
 
 5. Identify a core journal list/ secondary journal list. 
 
 6. Compare lists with your own holdings to determine how well the core 
 journals and secondary journals are represented. 
 
 This approach will identify the core journals in a specific subject disci- 
 pline. This is an abstract list in that it will not necessarily correspond to 
 users' needs. BRS has taken this approach one step further and provides a 
 
752 JANE L. CALDWELL 
 
 special service to assist managers in collecting relevant data for quantita- 
 tive collection analysis. BRS will save all citations that are requested from 
 offline print requests for one calendar year. These offline prints are the 
 ones that have been generated in response to users' queries and should 
 represent the needs of actual users. Because users often use online services 
 when looking for information on new or "hot" topics, the collection 
 development service will represent current demands on the literature. BRS 
 will save the journal citations and generate an annual or semiannual 
 report, depending upon the service option that was chosen. The report, 
 broken down by database, will consist of two sections: (1) an alphabetical 
 listing by journal title; and (2) a rank order by frequency of citation. 
 
 In my former life (before joining BRS) I was at the University of 
 Minnesota's Education/Psychology/Library Science (EPLS) library. One 
 of our continuing projects was the evaluation of our collection and we used 
 the BRS collection development service as one of the tools to judge the 
 strength of our journal collection and to identify specific titles that should 
 be considered for inclusion in the collection. This study was reported by 
 Celia Ellingson and Lori Hedstrom. Their methodology: 
 
 1 . Select databases that are primarily related to the collection and are used 
 enough to warrant further analysis. This limits the focus to ERIC, 
 Exceptional Child Educational Resources, National Clearinghouse for 
 Mental Health, and Psychological Abstracts. 
 
 2. Eliminate from study journal titles that were cited fewer than six times. 
 This threshold was perhaps arbitrary, but it relates to the fact that 
 interlibrary loan can be used for the first five requests. 
 
 3. Check each title with six or more citations for holdings information and 
 location within the University of Minnesota libraries. Journals that 
 were not located within the EPLS library, but had multiple campus 
 locations, were noted for their location most physically available to 
 to EPLS. 
 
 4. Count the total number of titles to determine the percentage held within 
 the EPLS library, held in other campus locations, and not held. 
 Example: Online serials list, like the California Union List of 
 Periodicals (GULP), could provide an even simpler alternative. 
 Example: 
 
 ERIC 
 
 Total 544 
 
 Held at EPLS 364 66.9% 
 
 Held elsewhere 93 17.0% } 83.9% 
 
 Not held 87 16.1% 
 
BIBLIOGRAPHIC DATABASES 753 
 
 NIMH 
 
 Total 66 
 
 HeldatEPLS 34 51.5% 
 
 Held elsewhere 25 37.8% } 89.3% 
 
 Not held 7 10.7% 
 
 5. Compile a merged list of all journals held outside the EPLS library, but 
 within the University of Minnesota (twin cities campus) libraries. 
 Example: 
 
 Architecture Library 
 
 Planning for Higher Education (21) 
 
 Progressive Architecture (14) 
 
 Environment and Behavior (10) 
 Biomedical Library 
 
 Journal of Autism & Childhood Schizophrenia (359) 
 
 American Journal of Psychiatry (145) 
 
 Journal of Medical Education (102) 
 
 British Journal of Psychiatry (90) 
 
 Behavior Therapy (80) 
 
 American Academy of Child Psychiatry Journal (71) 
 
 Journal of Nervous b Mental Disease (71) 
 
 Hospital & Community Psychiatry (67) 
 
 G/C/T(91) 
 
 Bureau Memorandum (56) 
 
 Pointer (56) 
 
 AAESPH Review (50) 
 
 Australian Journal of Mental Retardation (42) 
 
 Psychologic a Patopsychologie Dietata (33) 
 
 Child: Care, Health b Development (32) 
 
 Teaching at a Distance (30) 
 
 Special Education in Canada (29) 
 
 United States Air Force RHL Technical Report (28) 
 
 6. Compile a merged list of all journals that were not held by any 
 University of Minnesota library (checking holdings was a relatively 
 simple process because of the existence of the Minnesota Union List of 
 Serials. A student assistant was able to do the checking during his/her 
 spare time). 3 
 
 The University of Minnesota is one of the many libraries across the 
 country that has taken advantage of the BRS collection development 
 service. To date, most users have come from academic libraries, although 
 there would be many advantages for corporate libraries. 
 
JANE L. CALDWELL 
 
 Before I go on, I would like to review the assumption on which the 
 collection development service is based and that is that citations retrieved 
 in literature searches represent the actual demands of our clients. We have 
 to assume that our users take the journal literature and the literature search 
 seriously enough to want to read the articles that are retrieved. We have to 
 assume that the users of the literature search service are representative of all 
 our users. We also have to assume that the searches were relevant to the 
 subject and that the literature represented was actually about the desired 
 subject. 
 
 Surveys like the one from the University of Minnesota point out 
 several uses for the BRS collection development service. It can be used to: 
 
 1. Judge the relative strengths of holdings within a specific library or 
 group of libraries. 
 
 2. Identify core journal titles that should be acquired by a specific library. 
 
 3. Identify low-demand titles that could be cancelled if necessary. 
 
 4. Compare the requests of our users with other types of surveys that 
 identify core literature in a specific field. 
 
 The collection development service provides a quantitative measure 
 of potential use of a library collection. It is not meant to be used without 
 the professional judgment of the librarian or subject bibliographer. There 
 are, of course, other factors to consider i.e., the service measures only the 
 requests from users of a literature search service. In libraries that charge for 
 access to the service, these statistics would not represent needs of all users. 
 In an academic library, undergraduate students would not be as well 
 represented as graduate students and faculty. Rich departments that could 
 afford more literature searches might be over represented. And, most 
 importantly, the service is valid only in disciplines where online databases 
 adequately cover the published journal literature and where journal publi- 
 cation is the accepted means of communication between scholars. The 
 collection development service provides a quantitative measure of poten- 
 tial user demand and, when used in conjunction with qualitative mea- 
 sures, provides a rational method of collection evaluation. 
 
 Another online tool that can be used for a bibliometric study is the 
 overlap study. Several years ago, I was involved in a study the goal of which 
 was to compare overlapping retrieval between the ERIC and Psychological 
 Abstracts databases. Before our study, we knew that there should be a high 
 percentage of overlap between the two files in subject areas such as 
 educational psychology which were equally represented in both data- 
 bases. Our findings, published in the journal Database, June 1979, were 
 surprising. 5 The percentage of citations that were duplicated. ..was less 
 than might be expected based upon the percentage of overlap in journal 
 coverage between the two databases...." After comparing the journal titles 
 
BIBLIOGRAPHIC DATABASES 755 
 
 produced from the retrieved citations with the list of journals the database 
 producers claimed to index, we discovered that "more than twice the 
 number of citations actually duplicated could have appeared as dupli- 
 cates." Our findings in the case of ERIC and Psychological Abstracts 
 indicated that for these two databases, a search in a cross-disciplinary topic 
 warranted a multiple file search. 
 
 At the time of our study, it was necessary to do a time-consuming 
 comparison between the results of the several online searches. BRS now 
 offers a unique feature MERGE that allows the searcher to merge 
 together searches from multiple files. Duplicate citations can be easily 
 revealed, and in addition to the obvious convenience offered to the con- 
 sumer of the search, the MERGE feature offers valuable information on 
 overlap between databases. 
 
 Often, it is a management decision to set policy on whether multiple 
 files are routinely accessed to answer specific queries. The merge feature 
 can assist managers, after studying results of various searches, to determine 
 whether multi-file searches are necessary. In most library settings, it is 
 often necessary to elect the most cost-effective single database search. 
 Overlap studies and use of the BRS merge feature can be very useful to 
 justify these decisions. 
 
 I have discussed uses of online databases to evaluate journal collec- 
 tions and the content of various indexes. There are several other nontradi- 
 tional uses of online databases that can provide insights into the research 
 habits of your clientele and thus better determine priorities of library 
 service. 
 
 It is possible to use online databases to discover the research interests 
 and citation habits of your clientele. Author searching can give you the 
 subject specialities of your clientele, but by examining their citation habits 
 through the use of a citation index like Science Citation Index or Social 
 Sciences Citation Index you can unearth their ongoing research interests, 
 the journals that they regularly scan (and probably subscribe to), and other 
 authors that they favor. It is also possible to see if a researcher cites the same 
 papers over and over again, or if he is adventurous and strays from his 
 standard sources. It is then possible to speculate as to whether your library 
 is really used! 
 
 Trends in publication can be determined by use of databases like 
 Books in Print, Brodart's Booksinfo file and Ulrichs. It is possible to pull 
 up all books recently published by a particular publisher to scan quickly 
 their current output, or to take the opposite approach and do a subject 
 search and determine which publishers favor particular subjects. Price 
 information is searchable and within the capability of BRS files so it is also 
 possible to use online databases to determine the average price of a book in 
 a particular subject area, to determine what percentage of books cost more 
 
156 JANE L. CALDWELL 
 
 than a specific price, etc. It is possible to use online databases to determine 
 how much literature exists in a topic before adding a new program or 
 course in an academic institution. Online searching can provide the 
 decision-maker with a starting point for justification of a book budget. 
 
 If the researcher cites papers from journals held by your library and 
 not readily available elsewhere, or cites journals that you know he has 
 requested or that have appeared in a computerized literature search, you 
 can be relatively sure that the researcher has been using your library 
 services. 
 
 To identify the research interests and publication output of an organi- 
 zation, search under the institution name to determine its publication 
 output, see in what format it usually publishes, its subject area strengths, 
 and related interests. Searching online is often the only way that institu- 
 tion names are accessible and it provides a quick and easy way to measure 
 an organization's research activity. Do not forget to use files that cover 
 patent information often patents are the best measure of ongoing 
 research in technical fields. 
 
 You can determine the "invisible college," or group of scholars who 
 communicte directly with each other, by using cocitation analysis. Cocita- 
 tion analysis is a sophisticated technique. For the purpose of discussion, 
 we will simplify it to an examination of which authors cite each other. 
 Because these cocitations are often based upon prepublication correspon- 
 dence between experts, it is possible to speculate upon what core of 
 researchers form an "invisible college." Computerized searching can make 
 it possible to crack this core of expertise. Databases like Science Citation 
 Index and Social Sciences Citation Index can be used to determine cocita- 
 tion. Databases like American Men and Women of Science can be searched 
 to discover experts in a specific field. 
 
 A different, and sometimes controversial, use of online databases is to 
 use them to gather supporting evidence for faculty promotion and tenure 
 decisions. Online databases provide a cost-effective method to quickly pull 
 together complete bibliographies of all members of a department, whether 
 in a university or other research organization. Individual productivity can 
 be measured. The strengths of an entire department or organization can be 
 evaluated and compared with similar departments or organizations. Data- 
 bases that index material other than periodical literature, such as ERIC, 
 NTIS, DOED, or SSIE can show evidence of the ability to successfully 
 compete for research grants. 
 
 Citation indexes provide a way to determine the relative value placed 
 upon the research produced by one's own faculty or researchers. Relative 
 rankings or accreditation surveys of departments or of graduates of depart- 
 ments can be compiled by examining how frequently the department 
 faculty or graduates are cited. Although it is possible to produce similar 
 
BIBLIOGRAPHIC DATABASES 757 
 
 evaluation studies manually, online searching makes this type of survey 
 economically feasible. 
 
 I have discussed how various types of "bibliometric" studies can be 
 used to provide management information in libraries. Another aspect of 
 online database searching is that it provides a sometimes overwhelming 
 amount of information to library managers in the form of statistics gener- 
 ated by the online search service. The statistics, a byproduct of the service, 
 provide information on who uses one's services and why they use them. 
 The search statistics can provide cost figures how much an average 
 search costs in connect time, print charges, and personnel time that can 
 help the manager justify a budget. The annual increase in search volume 
 can help one argue for an increase in staff or at least justify the present 
 staff. Sometimes it is necessary to use search statistics to determine which 
 search analysts do not make cost-effective use of online time perhaps it is 
 necessary to invest in additional training. 
 
 I would like to review some of the information that should go into a 
 search log, and ways one can use that information. 
 
 1. Search number. There are two different theories on how a search 
 should be numbered. One theory is that a search should be a single 
 intellectual query. It does not matter how many databases are accessed 
 to answer the question. (The analogy would be that in traditional ref- 
 erence services you do not count each printed tool consulted as a 
 separate search.) Another method is to count access to each database as 
 a search. This is a better representation of the amount of work 
 that is done, but it can be very misleading. Some institutions count 
 access to different online segments of the same database as separate 
 searches. A sensible compromise would be to keep counts of both. 
 
 2. Search requester. This could be keyed to keep track of department 
 use and status of each individual user. 
 
 3. Purpose of search. Is this search to answer a ready reference question 
 or to compile an extensive bibliography? Is the requester asking for 
 information to write a short paper, to give a talk next week, or to 
 review the literature prior to starting a dissertation? 
 
 4. Who will be charged for the search? Sometimes it is advantageous to be 
 able to go back to your records to determine what departments have 
 been using your services. You might need their support later! Another 
 key point, if you are charging for your services, is to determine the level 
 of support coming out of your users own pocketbooks. 
 
 5. Search question. A brief statement will do; it is helpful to code for 
 broad subject categories. 
 
 6. Databases accessed. 
 
 7. Vendors used. 
 
158 JANE L. CALDWELL 
 
 8. Prints both online and offline. 
 
 9. Sign on and sign off time for each database. 
 
 10. Calendar day. 
 
 11. Elapsed time. 
 
 12. Estimated charges. 
 
 13. The name of a search analyst. 
 
 When the search comes back, one will want to add the actual charge and 
 after the search analyst and requester have looked over the search, one may 
 want to add evaluative comments. 
 
 Online search statistics, like statistics generated from traditional 
 library services, provide hard data for library managers to use in evaluating 
 their services and budgeting for the future. The difference between tradi- 
 tional reference services and online services lies in the fact, unfortunately 
 sometimes, that a bill is generated each time we go online. For internal 
 accounting procedures, we need to justify our invoice to actual services 
 provided. This produces accurate record keeping because we are forced to 
 log in parts of the reference process that are normally not recorded. Thus 
 we can better judge the time it takes to handle a specific question and judge 
 the costs of providing information. 
 
 BRS provides an online accounts file that helps searchers to tailor the 
 search service to fit their specific needs and provides managers with the 
 most up-to-date figures on online usage. The accounts file will display 
 usage on specific files or access to all the BRS databases. The online usage 
 will be broken down by current month, last month and the year up to last 
 month. One will always have access to information that will help deter- 
 mine the online usage and expenses. 
 
 BRS, always trying to meet the needs of its users, is now developing an 
 online record keeping system. It will provide a way for one to store this 
 management information the same information that would go into a 
 search log. Mondily reports could be generated that will indicate how one 
 is using online services, who is using the service and for what purpose, 
 average time and cost per search, what databases are being searched, etc. 
 Many libraries that have access to their own mini- or microcomputer 
 already have the means to store and manipulate this type of information. 
 The statistics online file will give all libraries that advantage. 
 
 I have discussed ways that online searching can provide one with a 
 wealth of management information. Much of the information provided is 
 quantitative in nature. This is not a negative feature for we need objective 
 measures to evaluate our services. However, quantitative surveys should 
 always be used in conjunction with concern for quality as well as quantity, 
 and with the measured judgment of the trained librarian. 
 
BIBLIOGRAPHIC DATABASES 159 
 
 REFERENCES 
 
 1. Hawkins, Donald T. "Unconventional Uses of On-line Information Retrieval 
 Systems: On-line Bibliometric Studies." JASIS 28(Jan. 1977): 13. 
 
 2. Caldwell, Jane L., and Ellingson, Celia S. "Online Bibliometrics." BRS Brief Paper 
 Series no. 4, June 1980, p. 1. 
 
 3. Ellingson, Celia S., and Hedstrom, Lori A. "Using Online Databases as a Tool for 
 Collection Development." BRS Brief Paper Series no. 15, May 1981, p. 3. 
 
 4. Caldwell, Jane L., and Ellingson, Celia S. "A Comparison of Overlap: ERIC and 
 Psychological Abstracts." Database 2(June 1979):66-67. 
 
 5. Ibid., p. 66. 
 
 BIBLIOGRAPHY 
 
 Broadus, Robert N. "The Applications of Citation Analyis to Library Collection Building." 
 
 In Advances in Librarianship, vol. 7, edited by Melvin J. Voight, and Mk hael H. Harris, 
 
 p. 299-335. New York: Academic Press, 1977. 
 Caldwell, Jane L., and Ellingson, Celia S. "Online Bibliometrics." BRS Brief Paper Series, 
 
 no. 4, June 1980. 
 Garfield, Eugene. "Citation Analysis as a Tool in Journal Evaluation." Science 178(3 Nov. 
 
 1972):471-79. 
 Goffman, William, and Morris, Thomas G. "Bradford's Law and Library Acquisitions." 
 
 Nature 226(6 June 1970):922-23. 
 Halperin, Michael. "Preparing Online Service Reports with Packaged Programs." Online 
 
 5(Oct. 1981):62-68. 
 Line, Maurice B., and Sandison, Alexander. "Practical Interpretation of Citation and Library 
 
 Use Studies." College 6- Research Libraries 36(July 1975):393-96. 
 Narin, Francis, and Moll, Joy K. "Bibliometrics." In Annual Review of Information Science 
 
 and Technology, vol. 12, edited by Martha Williams, pp. 35-58. White Plains, N.Y.: 
 
 Knowledge Industry Publications, 1977. 
 
EDWIN B. BROWNRIGG 
 
 Director 
 
 Division of Library Automation 
 University of California at Berkeley 
 
 An Online General Ledger System 
 
 Across academia, decision-makers with small to large budgets often have 
 only the fuzziest idea of their actual budget balance, or what their balance 
 might look like at some point in the future. Somehow, this situation 
 persists with decision-makers who have an uneasy reliance on some kind of 
 accounting process (sometimes automated), which generally they do not 
 control directly. Typically, a manager's dependence on the so-called 
 "accounting department" introduces an information lag of several weeks 
 or even months. Thus, managers look at their budgets through a rear view 
 mirror; and projecting their actual budgets into the future becomes an 
 exercise in divination. 
 
 In January 1979, the University of California Division of Library 
 Automation (DLA) found itself with the problem of maintaining a timely, 
 ledger. Also at that time, DLA was building a large-scale IBM operating 
 system to support the University of California Online Union Catalog. 
 This system also had all of the computing resources to support a 
 sophisticated general ledger system. 
 
 So, an obvious issue emerged as to whether it was more cost-effective to 
 buy, or to write from scratch, a general ledger program. Common sense 
 suggested that it ought to be simple to buy such software. However, 
 exhaustive investigation revealed that flexible general ledger systems came 
 as part of a turnkey hardware/software solution. The fact was that the 
 IBM/370 operating system proprietary software was either mostly batch, or 
 otherwise required costly teleprocessing software which was not a part of 
 DLA's intended software inventory. Timesharing was considered too, but 
 even if the right software could have been identified and located, its cost 
 would have been greater than writing the same programs in-house. The 
 
 160 
 
ONLINE GENERAL LEDGER SYSTEM 161 
 
 unexpected conclusion was that it was cost-justifiable to devote a 
 programmer's time to develop an online general ledger system that would 
 be tailor made to DLA's needs. 
 
 Nature of the Application 
 
 The essence of maintaining a general ledger is to create budget lines, 
 to encumber and to expend against them, to have the balances of each line 
 adjusted automatically, and to carry updated subtotals as well as a grand 
 total. For example, if a general ledger were to have a line for 
 "telecommunications," then its online implementation should support 
 the following general functions: 
 
 Associate the line entitled "telecommunications" with a fund code. 
 
 Allocate an amount of money for "telecommunications." 
 
 Adjust the allocation any time in the future. 
 
 Encumber funds against "telecommunications." 
 
 Adjust an encumbrance against "telecommunications" any time in the 
 future. 
 
 Allow full or partial expenditures against a particular encumbrance 
 against "telecommunications." 
 
 Allow expenditures from "telecommunications" for which an encum- 
 brance was not made. 
 
 Maintain a balance for "telecommunications," while adjusting the 
 subtotal, of which "telecommunications" is a part, as well as the grand 
 total. 
 
 Details of Implementation 
 
 The general ledger system developed at DLA has no name. DLA runs 
 it under IBM's Time Sharing Option (TSO), and in theory it will run 
 under any IBM/370 operating system. TSO handles communications with 
 the terminal, but it could also be accommodated by CMS, CICS, etc. The 
 application proper is written in PL/1. All of the programs were written at 
 DLA by Michael Thwaites. File management is through sequential 
 (QSAM) and basic direct (BDAM) access. For database management activi- 
 ties like indexing, much advantage is made of large regions of real 
 memory. All transactions are check-pointed for restart. For simplicity and 
 to keep costs down, dumb ASCII terminals are used in line-by-line mode. 
 Through simple screen clearing, cursor addressing and a data rate of 1200 
 bits per second, a very friendly software interface eases the user through 
 learning and use of the system. 
 
162 EDWIN B. BROWNR1GG 
 
 Design Criteria 
 
 The design of the system takes into serious account the fact that the 
 person who actually plugs numbers into it will want to do so without 
 exhaustive training, but with ease, confidence and a minimum of time at 
 the terminal. To those ends, careful attention was paid not only to the 
 textbook functions of the general ledger, but also to the idiosyncrasies of 
 the purchasing and accounting offices through which DLA operates, and 
 to the preferences of those responsible for both the broad management and 
 hands-on use of the system. 
 
 Detailed performance specifications were developed by a systems anal- 
 ysis team that included those mentioned earlier the director and the 
 programmer. The result was a printed document of some sixty-two pages 
 that became the basis for the operator's manual. 
 
 Logging onto the system is a function that embodies many exemplary 
 features of the system: brevity, ease of use, security, and selectivity. Within 
 ten lines of text 99 percent of which is coming from the computer not the 
 user a session is provisionally established, most recent activity is 
 recounted, the user enters the appropriate passwords, and limits of privi- 
 lege are established based on the user's alleged identity. 
 
 In general, there are three levels of privilege: absolute authority to read 
 and write anything (such as setting up fund codes, budget titles, grouping 
 lines for subtotaling, and patching any number in the ledger). The next 
 level of privilege is to encumber and expend. The least privileged users can 
 only read the ledger. 
 
 The user communicates to the online general ledger in a command 
 language. The number of functions is very finite and the command verbs 
 are highly suggestive. To aid the user's memory, the system will issue an 
 appropriate help message whenever asked to. 
 
 Because the system uses line-by-line terminals, editing is very simple. 
 There are no function keys. The system allows the user to review a transac- 
 tion in its entirety before executing it. Even if the user makes a mistake, it 
 can be undone. 
 
 The online general ledger functions so as to reduce as much as possi- 
 ble problems with the mechanical aspects of the system. Not only does it do 
 all of the arithmetic, but it also helps the user to verify that the right vendor 
 is paid, and that only one payment is made to that vendor. It even allows 
 credits and discounts to be posted to a particular purchase order to invoice. 
 Output from the online general ledger can go to a terminal or be spooled to 
 a printing device. When used in conjunction with an IBM 6670 or a Xerox 
 9700 online laser printer, the hardcopy output is particularly pleasing. 
 
ONLINE GENERAL LEDGER SYSTEM 163 
 
 Limitations 
 
 One of the design criteria for the online general ledger was to limit its 
 implementation exclusively for DLA. Nevertheless, other departments at 
 the university can use it for internal ledger management. It cannot be used 
 in a distributed fashion where several departmental ledgers might be 
 aggregated, and that was a conscious design limitation. 
 
 Conclusion 
 
 The experience at DLA with a home-built general ledger system has 
 been very positive. The director can look at the budget from his terminal 
 any time he wants to, and see it up to the moment of the keying of the most 
 recent purchase order and invoice information, which takes place virtually 
 every day. It takes one person about twenty minutes each day to keep up 
 with keyboarding. Since DLA has its own computer center, this system 
 runs for no recurring cost of its own. But most importantly, it provides 
 satisfaction with a bare minimum of human involvement. 
 
GLYN T. EVANS 
 
 Assistant Vice-Chancellor for Library Services 
 
 State University of New York 
 
 Central Adminstration 
 
 ALBERT BEILBY 
 
 Acting Director of Education and Training 
 
 Training Department 
 Binghamton Psychiatric Center 
 
 A Library Management Information System 
 in a Multi-Campus Environment 
 
 Introduction 
 
 The Office of Library Services in the Central Administration of the State 
 University of New York (SUNY) has, since 1975, been developing a library 
 management information system based on the analysis of library and other 
 bibliographic and academic data which are available in machine readable 
 form. Although primarily designed for the SUNY libraries, the processes 
 are applicable in other academic libraries because of the general availabil- 
 ity of the data used in the system. The task has changed over the years as 
 new ideas and opportunities were realized, as new appreciations of the 
 obtained results were attained, and as the technical environment has 
 evolved. Nonetheless, the fundamental structure of the system design has 
 not changed since the first ideas in 1974. 
 
 This is an interim report. Progress has been agonizingly slow for two 
 reasons. First, the difficulty of obtaining support and resources has been a 
 real hindrance; the work has been squeezed into overcrowded schedules 
 and ever-straitening budgets. Second, many of the machine-readable data 
 which one confidently felt would be available in the late 1970s or very early 
 1980s are still not available. Some years, at least, will pass before the work 
 can be completed as we see it now. Who knows what new ideas and 
 opportunities will emerge as new results become available? Nonetheless, 
 enough has been achieved to justify this report. 
 
 164 
 
MIS: MULTI-CAMPUS ENVIRONMENT 165 
 
 Environment 
 
 The State University of New York is a multi-campus university in 
 New York State. It is composed of thirty-two state operated campuses and 
 thirty-two community colleges which are administered by local county 
 authorities. The senior colleges in New York City and ten community 
 colleges in the city compose City University of New York (CUNY), a 
 separate organization. 
 
 The total student head count enrolled at the state operated campuses 
 (i.e., excluding community colleges) in 1982 was 385,000, with 29,200 
 faculty and staff. The total collection size is slightly more than 10.7 million 
 cataloged volumes, growing at the rate of some 400,000 volumes per year. 
 The acquisitions budget in 1982 was slightly more than $12.8 million. 
 
 There are thirty-five separately administered and budgeted libraries 
 within the SUNY state-operated campuses the primary locus of these 
 studies. They are composed of four university centers (including a law 
 school library), four medical schools, twelve four-year colleges of arts and 
 sciences, six two-year agricultural and technical colleges, four special 
 colleges (forestry, maritime, optometry, and technology), and four statu- 
 tory colleges (Alfred Ceramics, Cornell Agriculture and Human Ecology, 
 Industrial Labor Relations and Veterinary Medicine libraries). All the 
 variety one's heart could desire. 
 
 The Office of Library Services in the Central Administration of SUNY 
 is charged with planning, developing and integrating the library resources 
 of SUNY in support of its academic programs. In an early step to achieve 
 this goal, the office contracted with OCLC in 1973 to provide services to 
 State University and other New York participating libraries. As a result, 
 the SUNY/OCLC Network is also administered by the Office of Library 
 Services. The network now comprises 228 institutions (academic, public, 
 school, law, medical, state agency, etc.) and for-profit institution libraries. 
 A further 600 (approximately) are either sharing institutions or are 
 members of processing centers or Regional Union Lists of Serials. Within 
 the state, 70 percent of the independent higher educational institutions, 82 
 percent of public higher education, and 78 percent of the public library 
 systems participate in the network. 
 
 As one of its services, the SUNY/OCLC Network stores and processes 
 OCLC distribution tape records for its libraries at the SUNY Central 
 Administration Computer Center. Currently over 10 million records are 
 housed for the libraries, with the file growing by approximately 2 million 
 each year. There is a clear relationship between this activity and the 
 development of the library management information system. 
 
 It should be stressed that the Central Office of Library Services does 
 not have responsibility for the direct operation of the campus libraries, 
 
166 GLYN T. EVANS 6- ALBERT BEILBY 
 
 which report ultimately to each campus president. The central office has a 
 planning function in which it tries to create as hospitable an environment 
 as possible for the campus libraries. The office is guided by Stafford Beer's 
 dictum that "the only feasible ultimate objective of systemic control is to 
 hold the system within its natural boundaries." 1 
 
 Decision Support Systems 
 
 The computer has been seen as an aid to management from the time of 
 its adoption by organizations as an administrative tool (as opposed to a 
 research or production tool). It has been expected that data which emerged 
 as a byproduct of (or could be coaxed from) a production operation would 
 be collected, collated and analyzed to provide management with data 
 which would improve the operation of the system. In library circles, 
 circulation data seems to attract the most attention. It must be said, 
 however, that recorded examples of such data actually being used to make 
 decisions are extremely rare. A paper (such as that of F.H. Spaulding and 
 R.O. Stanton from Bell Laboratories 2 ) which records the effect circulation 
 had on acquisition/selection decisions is a desert rose. Indeed, the develop- 
 ment of management information systems seems to have created yet 
 anodier battleground for internal control of any organization well de- 
 scribed by Peter Keen in his article "Information Systems and Organiza- 
 tional Change." 3 
 
 The term and activity of management information systems is now 
 becoming supplanted by decision support systems (DSS), the fundamental 
 difference being that instead of providing passive displays of data (proba- 
 bly offline), a true decision support system goes one step further. The data 
 are available online in a synthesized form from a variety of sources, and are 
 presented to the administrator through a "friendly" terminal which sup- 
 ports modeling programs, color graphics and other facilities all of which 
 give the manager the opportunity to review the data and to test alternate 
 strategies. King defines a decision support system as, "a computer-based 
 system that the administrator uses to amplify or improve judgement. It is 
 not a system that makes decisions." 4 DSS software has the following 
 capabilities: (1) report preparation and inquiry; (2) modeling language; 
 (3) graphic displays; and (4) financial and statistical routines. Hopkins 
 and Massy write: 
 
 The process of modelling is always one of synthesizing known facts, 
 theories, and judgements into a meaningful pattern. 
 
 Models are about something; they purport to represent an aspect of 
 something that exists, or might exist, in the real world. We call the object 
 of a model the reference system. Thus a given reference system can in 
 principle be represented by many different models, each one more-or-less 
 
MIS: MULTI-CAMPUS ENVIRONMENT 167 
 
 accurate with respect to certain characteristics of the system.... Models 
 need to be verified. ..[and] validated. ...Models are designed for a purpose. 5 
 
 They also caution readers in a manner similar to King: "A good model 
 one that is simple but complete, stable yet adaptable should make the 
 quantifiable dimension of decision -making a far less mysterious place in 
 which to operate, but it will not thereby lessen the burden of choice." 6 
 The task at SUNY has been to build such a model, although in truth 
 the work began before the term and the attractions of the DSS were 
 developed. Some results have been obtained from the work to date, 
 although the project is not yet complete. DSS systems hold the promise of 
 reducing development time. 
 
 Stimuli 
 
 The SUNY Office of Library Services did not begin this work on an 
 idle whim. The libraries of the university were under severe fiscal 
 pressure and we had little idea then that the economic conditions would 
 continue to deteriorate as they have. Since other institutions have faced 
 some of the pressures, it is useful to list those seen as critical at SUNY: 
 
 1. Acquisition formula budgeting. The university had adopted the 
 Clapp-Jordan formula in 1968 and was using the formula to build the 
 collection. In 1975, the formula was used by the Division of the Budget 
 to cut back the acquisitions budgets for three of the university centers on 
 the grounds that the libraries were, or soon would be "adequate" 
 according to the formula. The loss of funds was two-thirds of $1 million 
 from the annual acquisition base. Further, a derivative of that for- 
 mula was being promulgated by the State Education Department as a 
 state-wide guideline. 8 These actions were clearly not in the best interests 
 of the university's libraries. The heat was on to develop another 
 formula. As it happens, work had already begun on a discipline-based 
 "formula," but was not complete in time to be offered as an alternative 
 to the cuts in funds. (It is of interest that the results of a project by Evans, 
 Beilby and Gifford completed in the process of developing an 
 acquisitions formula, concluded that it is not possible to derive a 
 "formula.") 9 It is possible to develop an information system which will 
 reflect the bibliographic components of the academic mission of a 
 campus. 
 
 2. Lack of an adequate statistical database. Apart from gross budgeting 
 data, and biennial Library General Information Survey (LIBGIS) 
 reports, 10 there was no firm database to describe the libraries, their 
 collections, their successes, or their failures. 
 
168 GLYN T. EVANS b ALBERT BEILBY 
 
 3. Isolation of library data from academic administrative data. Although 
 library data were gathered from LIBGIS, and summarized from budget 
 data, there was no link to relate them to academic program data such as 
 the location and enrollment in courses and programs, level of programs 
 (i.e., undergraduate or postgraduate) or location of programs. The 
 chicken and egg question of the location of programs and location of 
 library collections and which comes first could barely be 
 asked, let alone answered. The one reliable element was dollars ex- 
 pended per full-time equivalent (FTE) student, achieved by dividing 
 one number into another. 
 
 4. Traditional library data emphasis on "inputs" rather than "outputs." 
 Apart from total circulation, and interlibrary loan (ILL) traffic, there 
 was almost no emphasis on the collection of service data, which for a 
 service organization is incomprehensible. (Reference statistics were 
 added later.) LIBGIS surveys and the state guidelines emphasized space, 
 facilities, staff, collection size, and collection growth, rather than 
 performance. Not that those data are not important, they are. But one 
 wants to know how successful, or useless, one's library is; its service, not 
 its potential for service.) Furthermore, there was no evidence that the 
 "official" requirements for statistical reports would change as 
 machine-readable data became more available. 
 
 5. The multi-campus environment. Because of the multi-campus nature 
 of the university, and because of the nature of the growing ties among 
 libraries sharing the OCLC Network, it was clear from the beginning 
 that any system design must accommodate that variety and that added 
 dimension. Therefore, the system was designed from the ground up 
 with multiple campuses and multiple academic programs and libraries 
 in mind. Fortunately the program was prescient in that regard as fiscal 
 crises have begun to force the "trades and affiliations" of academic 
 programs within the university. 
 
 6. Fiscal pressures. The budgetary problems have already been cited. But 
 there are other, subtle factors which should be drawn into account: the 
 shift of monies to serials rather than monographs; new physical media; 
 the growing necessity to purchase information-on-demand through 
 database searches or ILL (as opposed to buying the potential to 
 supply service through acquisitions); or the decision to retain or discard 
 an item. 
 
 7. Political pressures. As would be expected, a sharp reduction in the 
 acquisitions budget tended to attract attention to the problems with 
 demands for a quick solution despite the absence of data. 
 
 A recitation of the earlier stimuli should not be interpreted as a 
 criticism of the university or the profession. Rather it is criticism of the 
 
MIS: MULTI-CAMPUS ENVIRONMENT 169 
 
 conventional wisdom, and the reluctance to accept both the need for data, 
 and the need to absorb them into operational decisions. It takes an uncon- 
 scionable amount of time and effort to effect change. 
 
 Given these stimuli, and the growing availability of machine-readable 
 data, it was decided to build a library management information system 
 which would satisfy the following purposes: 
 
 1. to establish a model that would describe the acquisition/retention 
 process in a multi-campus academic environment; 
 
 2. to develop and refine the available databases for inclusion in the 
 model i.e., to rely on data which have been acquired as the byproduct 
 of a production operation; 
 
 3. to establish computer programs which would drive the model and 
 provide reports; 
 
 4. to provide individual campus reports and system-wide 
 reports including time-series and trend reports; and 
 
 5. to integrate the reports into the planning processes of the libraries and 
 the institutions. 
 
 The uses to which the management information would be put include: 
 
 1. justification of acquisition budgets; 
 
 2. support of the planning process, particularly among: 
 
 (a) academic programs and library programs, 
 
 (b) campuses, and 
 
 (c)campuses and their local disciplinary environments; 
 
 3. provision of specific campus/interdisciplinary reports by library 
 disciplinary strengths and/or weaknesses; 
 
 4. support of the campus accreditation process; and 
 
 5. exploitation of the ability of the analytical programs to provide subject 
 and/or form bibliographies by discipline for a campus, or group of 
 campuses. 
 
 The Model 
 
 An early description of the model was reported in 1978, 11 but for the 
 benefit of continuity it is briefly described here. The structure is that of the 
 familiar five-box information system, comprising input, control, decision, 
 output, and feedback (see fig. 1). 
 
 The central decisions in a library is die acquisition and retention 
 decisions. The sum of these decisions is, in fact, the library. In an academic 
 environment, the inputs into that decision are supply i.e., what materials 
 are available and demand i.e., what academic and research programs 
 are supported by the library and for which community of users. The output 
 
C/3 
 
 oo 
 
MIS: MULTI-CAMPUS ENVIRONMENT 171 
 
 of the decisions is the library, and the catalog is the surrogate of the library. 
 The feedback into the decision (another source of input) is the use which is 
 made of the collection, defined by the discipline and the community, and 
 through the circulation, interlibrary loan and internal use. The control or 
 program element is the software and tables which drive the management 
 information system. 
 
 It was noted earlier that one of the criteria for the model design is that 
 the machine-readable data to be used would be operational data. Such 
 machine-readable data are available for each group described earlier 
 although not at present in all SUNY libraries. The data are summarized as 
 follows: 
 
 1. Input data (supply): Machine-readable files such as Bowker's 
 Books-in -Print, or American Book Publishing Record (ABPR). 
 
 2. Input data (demand): Campus enrollment data described with the U.S. 
 Office of Education, Higher Education General Information Survey 
 (HEGIS) disciplinary codes. 
 
 3. Ouput data: OCLC distribution tapes containing local library versions 
 and holdings of items cataloged into OCLC. 
 
 4. Feedback data: machine-readable files of circulation transactions from 
 automated circulation systems; and machine-readable files of ILL tran- 
 sactions from online network ILL systems. 
 
 It is shown that with the exception of the Input Demand HEGIS 
 Enrollment data, there is, or is likely to be, considerable consonance 
 among the data files. They all are bibliographic data files, and will 
 include, and transport, the same data elements among files depending 
 upon the purpose of the transaction. All library files (Bowker, OCLC, 
 circulation, and ILL) carry a Library of Congress (LC) classification 
 number, and probably an LC card number and an ISBN. Three of the files 
 carry, or can be made to carry, an OCLC number. 
 
 Files will also carry additional fixed field codes which assist in the 
 selection of data for analyses, and transaction codes which define the 
 nature of activity of the record itself. Examples of the latter are OCLC 
 update, produce or cancel codes. 
 
 The data elements which are used for analyses (as opposed to the 
 selection of records from a large file for analysis) in the segments of the 
 programmed model are the OCLC number and LC class number, used 
 singularly and in combination. Used by itself, the cooccurrence of the 
 OCLC number is a measure of the degree of overlap among collections. 
 The LC class number is an indicator of the subject strength of a library or a 
 group of libraries. Using both elements in conjunction, it is possible to 
 define both the collective subject strength of a group of libraries, and the 
 
772 GLYN T. EVANS 6- ALBERT BEILBY 
 
 degree of uniqueness or commonality of holdings of a library or all 
 libraries within each academic discipline. 
 
 As data become available following the installation of automated 
 circulation systems use, databased on LC class number and OCLC number 
 will be derived from ILL and circulation transaction files and entered into 
 the decision box as feedback data. 
 
 All of the discussed elements are well understood by library and 
 information professionals, but the academic administrative data used as 
 the "demand" segment of the input component of the model are less 
 familiar. The U.S. Office of Education's National Center for Educational 
 Statistics requires an annual HEGIS survey, through which all institu- 
 tions report the number of degrees awarded, the numbers and levels of 
 students and faculty, etc. To facilitate reporting, a disciplinary taxonomy 
 was established in 197 1. 12 SUNY has developed an automated statistical 
 reporting system which uses this taxonomy, plus fiscal information in a 
 Course and Section Analysis (CASA) file, to produce annual statistical 
 abstracts on trends and costs within the university. 1 There is thus available 
 a massive file of machine-readable data on the potential demand for library 
 services by discipline and by the university community. 
 
 The immediate and obvious problem is that the HEGIS/CASA file 
 does not carry any bibliographic data elements. However, the problem was 
 overcome in the research project by Evans, Beilby and Gifford noted earlier 
 which built conversion tables in which each term in the HEGIS taxonomy 
 is expressed as a series of LC class numbers, creating, in effect, a series of 
 mini-classifications. The ground rules are that ( 1 ) LC class numbers can be 
 drawn from any part of the class schedules, and (2) LC class numbers may 
 be used as many times as necessary. The HEGIS taxonomy has a two-tier 
 structure in which major classes are divided into subclasses (see fig. 2). For 
 its own statistical abstracts, SUNY has created a higher level amalgama- 
 tion of classes designated as disciplines. There are ten such groups. The 
 mechanism by which LC class terms can be assigned to a HEGIS subclass 
 and subsequently amalgamated into higher levels is indicated in figure 3. 
 
 After the structure was defined, individual library subject specialists 
 undertook to create the HEGIS/LC tables. It was found that over 13,000 
 LC classes were used to describe 494 HEGIS subclasses. (Figure 4 is a 
 sample of entries taken from the African studies HEGIS class. The descrip- 
 tions are taken from the LC class schedules.) 
 
 The use of LC class number, the HEGIS/LC tables, the OCLC record 
 number and the campus code (OCLC's three-letter symbol) allows the 
 identification of collection strengths and uniqueness related to teaching 
 demands at any campus or within a group of campuses. Or conversely, 
 their use can take a specific discipline or class and assess the relative 
 campus strengths in that area. Since each record which is assigned to any 
 
MIS: MULTI-CAMPUS ENVIRONMENT 173 
 
 04 BIOLOGICAL SCIENCES 
 
 0400 Biological Science Unclassified 
 
 0401 Biology General 
 
 0402 Botany General 
 
 0403 Bacteriology 
 
 0404 Plant Pathology 
 
 0405 Plant Pharmacology 
 
 0406 Plant Physiology 
 
 0407 Zoology General 
 
 0408 Pathology, Human, Animal 
 
 0409 Pharmacology, Human, Animal 
 
 0410 Physiology, Human, Animal 
 
 0411 Microbiology 
 
 0412 Anatomy 
 
 0413 Histology 
 
 0414 Biochemistry 
 
 0415 Biophysics 
 
 0416 Molecular Biology 
 
 0417 Cell Biology, Cytology 
 
 0499 
 
 Fig. 2. Example of the HEGIS Subclass Structure 
 
 class is individually identified, it is possible to create subject bibliogra- 
 phies by class to be used both as reference tools and for accreditation 
 assessment purposes. 
 
 The Process 
 
 In a simplified form, the following steps are taken to complete the 
 analysis: 
 
 1. receive OCLC tape; 
 
 2. read and extract the selected record use for analysis; 
 
 3. process the extract tape in the analysis program by (a) matching the 
 call with HEGIS/LC tables; (b) assigning it to levels (i.e., subclass, 
 class discipline, institution); (c) counting; (d) matching with CASA 
 
I 
 
 8 
 
 - 
 u 
 
 U 
 
 ~ 
 SO 
 
 
 00 
 
 ob 
 
MIS: MULTI-CAMPUS ENVIRONMENT 
 
 DT 730-995 
 G 2425-2739 
 G 8200-8903 
 GA 75-76 
 GA 286-286 
 GA 1341-1673 
 GB 330-378 
 GB 439-440 
 
 History, South Africa 
 
 Africa, Atlases 
 
 Africa, Maps 
 
 Africa, Mathematical Geography 
 
 Africa, Cartography 
 
 Africa, History of Map Production 
 
 Africa, Physical Geography 
 
 Africa, Geomorphology 
 
 GV 135-143 
 GV 1705-1713 
 HA 1951-2275 
 HC 501-591 
 HD 1169 
 
 Africa, Recreation 
 
 Africa, Dancing 
 
 Africa, Statistics 
 
 Africa, Economic History 
 
 Africa, Land Tenure 
 
 Fig. 4. Sample of Entries from the African Studies HEGIS Class 
 
 enrollment data specifically student credit hours per HEGIS class; 
 (e) performing overlap studies within each discipline at each level using 
 OCLC numbers; and (f) reporting by institution and level, listing titles 
 in brief entry of required listed titles. 
 
 Steps which are to be added to the process to complete the programming of 
 the model consist of: (1) incorporation of publishing data; (2) incorpora- 
 tion of use data; (3) development or acquisition of decision support system 
 software and evolution of system to an online interactive state. 
 
 The Results 
 
 Two analytic processes components of the total model are now 
 operational at the SUN Y Central Computing Center. They are component 
 analysis and overlap analysis. Component analysis is a process in which 
 catalog records from OCLC tapes are passed against the HEGIS/LC tables 
 
/ 76 GL YN T. EVANS b ALBER T BEILB Y 
 
 and allocated to subclasses, and the upward aggregates of class, discipline 
 and institution. The analyses may also be correlated with the number of 
 student credit hours taught in the discipline at the specific campus, based 
 on CASA file data. 
 
 Overlap analysis occurs when the cooccurrence of the same OCLC 
 number among the OCLC tapes of different campuses is used as a measure 
 of uniqueness and commonality of holdings, and the grades between (e.g., 
 held in three out of ten campuses). The overlap analyses are performed at 
 the institution level, or, following a component analysis, at subclass, class 
 or discipline level. 
 
 Results from these computer processes are designated as "obtained" 
 results. The obtained results may themselves be subject to subsequent 
 analysis, review and combination as indeed they would be in a decision 
 support system to generate "derived" results. A start has been made on 
 the process of producing derived results but by using the offline SPSS 
 (Statistical Package for the Social Sciences) or manual analyses. Both 
 obtained and derived results are reported in this paper. The derived results 
 are sufficiently valuable to justify the target of achieving operational access 
 to DSS software. 
 
 One major series of reports has been produced for the state university 
 campuses, with data derived from those studies being the basis of all the 
 results reported in this paper. In this series, the OCLC catalog tapes of 
 eleven colleges of arts and sciences (four-year colleges), for the period April 
 1977 to December 1979, most with a small percentage of graduate (masters) 
 programs were used. From this database, the study selected the latest use of 
 monograph records which had a transaction code of produce and an 
 imprint date of 1977 or 1978. This resulted in a base of 105,003 records for 
 analysis. The attempt clearly was to gain an understanding of current 
 acquisition decisions in the colleges. Other selections from the database 
 could have as easily been made, ranging from the whole database to serials, 
 updates and products. For our purpose, we chose the database we needed. 
 
 In the component analysis, the data were matched with student credit 
 hour (SCH) data from the CASA file for 1978 (for which data are collected 
 in the third week of the fall semester). We were exploring the academic 
 demand at the campus for the 1978-79 academic year, matched with the 
 acquisition of current (1977 or 1978) imprints which were received and 
 cataloged by the library between April 1977 and December 1979. The data 
 were analyzed first by the component analysis method, and then the 
 overlap analysis method at all levels. 
 
 Component Analysis-Obtained Results 
 
 Figure 5 is representative of a typical page from a computer printout 
 of the result of a component analysis. The four columns are respectively 
 
MIS: MULTI-CAMPUS ENVIRONMENT 177 
 
 the number of titles, the percent of titles allocated toeach HEGIS/LC class 
 for one campus from the database under review, the number of student 
 credit hours and the percent of student credit hours for that campus. 
 
 The reader is cautioned that the first two, and the last two columns are 
 added differently. The CASA/SCH data is a simple arithmetic sum. Either 
 the students are enrolled in, say, a three-credit hour course, or not. How- 
 ever, because of the ground rule in the creation of the HEGIS/LC tables 
 that it is possible for a class number to be assigned to more than one HEGIS 
 class, it is perfectly possible and reasonable for a specific title to be allo- 
 cated to more than one subclass in any one analysis. It is necessary for that 
 multiple allocation to be removed at each step of the upward aggregation 
 in order to avoid misleading and inflated results. Thus any multiple 
 allocation of a specific record title will show as supporting the subclasses 
 assigned, but will only contribute once to the class. A similar removal of 
 duplication occurs at the upward aggregations from class to discipline, 
 and discipline to institution. 
 
 This important point is illustrated in figure 5. The correct arithmetic 
 sum for the number of titles in the class "Letters" is 4594, yet the reported 
 number of titles is 4147. Given the consonance of the subclasses in the 
 group, it is not surprising that such a multiple allocation can occur. 
 Similarly, when the classes "Letters" and "Foreign Languages" are com- 
 bined into a single discipline, the reported number is 5049, but the arith- 
 metic sum (4147 + 1511) is 5658. This phenomenon also indicates an 
 important practical consideration. If it is possible to identify for any one 
 campus the subclasses and classes in which multiple allocation is taking 
 place, the books which are bought are obviously lower-risk investment 
 items than special areas of unique allocation a non trivial consideration 
 in times of fiscal crisis. Finally, in column three the percent of titles is 
 subject to the same rules of multiple allocation as the count of titles. 
 
 Component Analysis-Derived Results 
 
 One question raised by the component analysis is, simply: What is the 
 percentage allocation between the SCH and the current acquisitions, and 
 the ten-discipline HEGIS grouping? This allocation can be seen simply by 
 charting the ten points for each discipline on a graph containing both 
 acquisitions and the SCH. The results are demonstrated in figure 6. The 
 result, which was startling, is that the graphs for ten of the eleven campuses 
 were fundamentally the same as those shown by the three campuses in 
 figure 6. The eleventh campus, SUNY at Purchase the exception is a 
 new campus still busily building its basic collection. Thus the result is not 
 surprising. 
 
IS 
 
 t, w 
 
 O vC O -I O O 
 
 O O f- * 
 
 oc -J 
 m <N o 
 O 
 
 oooooo 
 
 O 00 . - 00 00 
 
 o o c> o o o*> ^> 
 
 (N (N 
 
 5. 
 
 a 
 
 1 
 
 a x oo 
 
 IB oo _ - _ : - 
 
 - CO X - 
 
 u 4, u >< a , -J 
 
 li V 
 
 . 
 
 O C 
 
 it c e - _ 
 
 U. J- ^ W! 
 
 -- & 
 
 c..= oocoecoob 
 T 
 
 e o -- 
 
10 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 c 
 
 M 
 
 
 
 ? 
 
 i 
 
 & 
 
 e 
 
 
 
 
 
 
 
 1 
 
 
 
 
 M 
 
 
 
 l 
 u 
 
 * 
 
 01 
 
 
 b 
 
 e 
 
 w 
 
 
 
 ec 
 
 
 < 
 
 B 
 
 
 
 
 
 c 
 
 
 ^ 
 
 wt 
 
 
 
 b 
 
 
 
 i 
 
 
 
 
 
 
 u 
 
 M 
 
 
 
 X 
 
 
 
 e 
 
 3 
 
 
 
 
 4 
 
 Ib 
 
 
 w 
 
 
 
 
 
 
 a 
 a 
 
 
 
 1 
 
 M 
 
 ^4 
 
 91 
 
 
 
 
 
 
 4 
 
 V 
 
 M 
 
 4 
 
 b 
 
 
 
 U 
 
 c 
 
 
 
 
 V 
 
 
 
 
 
 u 
 
 s 
 
 M 
 
 M 
 
 
 M 
 
 m 
 
 X 
 
 i 
 
 1 
 
 4 
 U. 
 
 3 
 
 I 
 
 
 
 I 
 
 "O QJ 
 C C 
 03 - 
 
 . 
 
 TJ - 
 
 2 
 
 
 c n 
 
 ^^ 
 
 gf 
 
 J 5 
 
 o 
 
 - 
 
 ( 3 
 
 (> . 
 
180 GLYN T. EVANS b ALBERT BEILBY 
 
 In another test, scatter-grams were performed to examine the possible 
 relationships between total credit hours and total titles acquired for each 
 HEGIS class. In a scatter-gram, the coincidence of the values for each 
 institution entered are displayed on the X and y axis. The more the 
 locations tend toward a straight line, the stronger are the relationships 
 between the elements. 
 
 Table 1 shows areas for which very strong relationships existed 
 between acquisitions and student credit hours. These eleven disciplines are 
 now described as "immanent" program areas, proposing that there is an 
 inherent relationship between the two elements. The third test undertaken 
 with component analyses data was to examine the degree of multiple 
 allocations among HEGIS classes, following the possibility of high v . low 
 risk acquisition investment. The higher the index of multiple allocation of 
 a title to more than one class, the higher the probability that the item will 
 be used. An index of less than one would indicate a low correlation between 
 the academic program and the acquisition program, diminishing as the 
 index decreases. Table 2 shows the indexes for eleven SUNY campuses. 
 
 Overlap Studies-Obtained Results 
 
 The overlap studies (see table 3) examine the cooccurrence of titles 
 among libraries based on the OCLC number at the subclass, class, disci- 
 pline, and institution levels. The programs are designed for a maximum of 
 ten institutions. This decision was frankly a programming compromise to 
 obtain results quickly by avoiding the delay caused by the complexity of 
 handling 100 institutions, as originally proposed. It has been found, 
 however, that there is so little overlap beyond ten institutions that there 
 may be little lost. 
 
 The results are displayed in a matrix in which the one column is the 
 individual institutions identified by their OCLC codes with the total titles 
 and copies in the last column. One column indicates the ten occurrences 
 from unique (i.e., held by that institution uniquely), two (i.e., the institu- 
 tion plus one other), to ten (i.e., held by all institutions). The final column 
 is the total of all titles. Each box in the matrix records the number and the 
 percentage in each column. Table 3 reports the overlap for one campus and 
 the total of all ten campuses among the subclass, class, discipline, and 
 institution level. The holdings of one campus will not necessarily follow 
 the pattern of the aggregate of all campuses. 
 
 The "total held by class" column is the sum of all copies held by a 
 particular distribution e.g., held in two or five libraries. The "actual 
 titles" column is the number of titles which overlapped. This is best 
 demonstrated in the ten-overlap column, where clearly the five titles held 
 by ten libraries in the subclass will yield fifty copies (see table 3). Figure 7 
 describes graphically the overlap at the institution level found in this series 
 of tests. 
 
MIS: MULTI-CAMPUS ENVIRONMENT 
 
 181 
 
 TABLE 1 
 
 MID-RANGE CLUSTERS AND THE PROGRAM AREAS 
 
 FOR WHICH STRONG RELATIONSHIPS WERE FOUND 
 
 BETWEEN ACQUISITIONS AND STUDENT CREDIT HOURS 
 
 Program Area 
 
 SUNY Colleges 
 
 SUNY Biological 
 Science & Health 
 Professions 
 
 0.80 @ .004 
 
 Brockport 
 Fredonia 
 Oswego 
 
 Buffalo 
 Geneseo 
 Potsdam 
 
 Cortland 
 New Paltz 
 
 Biological Science 
 
 0.97 @ .001 
 
 Buffalo 
 Geneseo 
 Platlsburgh 
 
 Cortland 
 New Paltz 
 Purchase 
 
 Fredonia 
 Oswego 
 
 Health Professions 
 
 0.98 @ <.001 
 
 Brockport 
 New Paltz 
 
 Fredonia 
 Potsdam 
 
 Geneseo 
 
 SUNY Fine 8c 
 Applied Arts 
 
 0.86 @ .007 
 
 Brockport 
 Geneseo 
 Plattsburgh 
 
 Cortland 
 New Paltz 
 Purchase 
 
 Fredonia 
 Oswego 
 
 Letters 
 
 0.76 @ .009 
 
 Buffalo 
 Geneseo 
 Plattsburgh 
 
 Cortland 
 New Paltz 
 Purchase 
 
 Fredonia 
 Oswego 
 
 SUNY Math & 
 Computer Science 
 
 0.91 @ <.001 
 
 Buffalo 
 New Paltz 
 Plattsburgh 
 
 Cortland 
 Oneonta 
 Purchase 
 
 Fredonia 
 Oswego 
 
 Mathematics 
 Physical Sciences 
 
 0.84 @ .009 
 0.93 @ .001 
 
 Buffalo 
 New Paltz 
 Purchase 
 Buffalo 
 New Paltz 
 
 Cortland 
 Oneonta 
 
 Cortland 
 Oneonta 
 
 Fredonia 
 Oswego 
 
 Fredonia 
 Oswego 
 
 SUNY Social 
 Sciences 
 
 0.87 @ .003 
 
 Cortland 
 Oneonta 
 
 Fredonia 
 Oswego 
 
 New Paltz 
 Purchase 
 
 Economics 
 
 0.88 @ .002 
 
 Buffalo 
 Oneonta 
 Purchase 
 
 Cortland 
 Oswego 
 
 New Paltz 
 Potsdam 
 
 Political Science 
 
 0.87 @ .006 
 
 Brockport 
 New Paltz 
 
 Buffalo 
 Oneonta 
 
 Cortland 
 Plattsburgh 
 
 *r Controlling for budget 
 
 It is one thing knowing the curve of the unique/commonality at the 
 level described thus far; but it does not tell a library specifically how it 
 relates to other libraries in the analysis. An extended report has been 
 defined in which any one library can assess its overlap title-by-title with the 
 
TABLE 2 
 
 INVESTMENT EFFICIENCY INDEXES FOR ELEVEN SUNY COLLEGES 
 OF ARTS AND SCIENCES 1977-78 IMPRINT DATE 
 
 Institution 
 
 TABLE 3 
 OVERLAP STUDIES 
 
 Index 
 
 Brockport 
 
 .86 
 
 Fredonia 
 
 .67 
 
 Buffalo 
 
 .81 
 
 Cortland 
 
 .83 
 
 Geneseo 
 
 .63 
 
 Oswego 
 
 .74 
 
 Plattsburgh 
 
 .77 
 
 Oneonta 
 
 .70 
 
 New Paltz 
 
 .75 
 
 Potsdam 
 
 .51 
 
 Purchase 
 
 .64 
 
 Mean 
 
 1.72 
 
 Institution 
 
 HEGIS Subclass-Afro American Studies 2211 
 Institution Total Held Actual 
 
 Code: XBM by Class Titles 
 
 Unique 23 
 17.42 
 
 68 
 9.48 
 
 68 
 33.66 
 
 2 21 
 15.90 
 
 72 
 10.04 
 
 36 
 17.82 
 
 3 16 
 12.12 
 
 54 
 7.53 
 
 18 
 8.91 
 
 4 10 
 7.57 
 
 48 
 6.69 
 
 12 
 5.94 
 
 5 14 
 10.60 
 
 85 
 11.85 
 
 17 
 8.41 
 
 6 11 
 8.33 
 
 72 
 10.04 
 
 12 
 5.94 
 
 7 12 
 9.09 
 
 91 
 12.69 
 
 13 
 6.43 
 
 8 11 
 8.33 
 
 96 
 13.38 
 
 12 
 5.94 
 
 9 9 
 6.81 
 
 81 
 11.29 
 
 9 
 4.45 
 
 10 5 
 3.78 
 
 50 
 6.97 
 
 5 
 2.47 
 
 Total 132 
 100.00 
 
 717 
 100.00 
 
 202 
 99.97 
 
TABLE 3 Continued 
 
 HEGIS Class-Social Science 22 
 
 . Institution Total Held Actual 
 Institution , VJJ ., , , _. . 
 Code: XBM by Class Titles 
 
 Unique 3,214 
 36.41 
 
 9,091 
 22.29 
 
 9,091 
 51.95 
 
 2 1,615 
 18.29 
 
 6,234 
 15.28 
 
 3,117 
 17.81 
 
 3 1,106 
 12.52 
 
 5,178 
 12.69 
 
 1,726 
 9.86 
 
 4 806 
 9.13 
 
 4,620 
 11.33 
 
 1,155 
 6.60 
 
 5 671 
 7.60 
 
 4,065 
 9.97 
 
 813 
 4.64 
 
 6 470 
 5.32 
 
 3,354 
 8.22 
 
 559 
 3.19 
 
 7 376 
 4.25 
 
 3,087 
 
 7.57 
 
 441 
 2.52 
 
 8 296 
 3.35 
 
 2,576 
 6.31 
 
 322 
 1.84 
 
 9 181 
 2.05 
 
 1,647 
 4.03 
 
 183 
 1.04 
 
 10 92 
 1.04 
 
 920 
 2.25 
 
 92 
 .52 
 
 Total 8,827 
 100.00 
 
 40,772 
 100.00 
 
 17,499 
 99.97 
 
TABLE 3 Continued 
 
 SUNY/CASA Discipline-Social Sciences 
 
 , Institution Total Held Actual 
 Institution r- A -vo\t j, /-/ T-.I 
 Code: XBM by Class Titles 
 
 Unique 4,228 
 37.88 
 
 11,262 
 
 22.97 
 
 11,262 
 52.52 
 
 2 2,176 
 19.49 
 
 7,850 
 16.01 
 
 3,925 
 18.30 
 
 3 1 ,356 
 12.15 
 
 6,222 
 12.69 
 
 2,074 
 9.67 
 
 4 965 
 8.64 
 
 5,452 
 11.12 
 
 1,363 
 6.35 
 
 5 798 
 7.15 
 
 4,825 
 9.84 
 
 965 
 4.50 
 
 6 545 
 4.88 
 
 3,912 
 7.97 
 
 652 
 3.04 
 
 7 439 
 3.93 
 
 3,598 
 7.33 
 
 514 
 2.39 
 
 8 336 
 3.01 
 
 2,920 
 5.95 
 
 365 
 1.70 
 
 9 213 
 1.90 
 
 1,953 
 3.98 
 
 217 
 1.01 
 
 10 103 
 .92 
 
 1,030 
 2.10 
 
 103 
 
 .48 
 
 Total 11,159 
 100.00 
 
 49,024 
 100.00 
 
 21,440 
 99.60 
 
TABLE 3 Continued 
 
 Institution-Ten SUNY Colleges 
 
 , Institution Total Held Actual 
 Institution vom* /-i 
 Code: XBM by Class Titles 
 
 Unique 8,241 
 37.80 
 
 25,474 
 24.26 
 
 25,474 
 53.88 
 
 2 4,167 
 19.11 
 
 17,256 
 16.43 
 
 8,628 
 18.25 
 
 3 2,642 
 12.11 
 
 13,281 
 12.64 
 
 4,427 
 9.36 
 
 4 1,938 
 8.88 
 
 11,776 
 11.21 
 
 2,944 
 6.22 
 
 5 1,521 
 6.97 
 
 10,035 
 9.55 
 
 2,007 
 4.24 
 
 6 1,166 
 5.34 
 
 8,700 
 8.28 
 
 1,450 
 3.06 
 
 7 884 
 4.05 
 
 7,224 
 6.87 
 
 1,032 
 2.18 
 
 8 666 
 3.05 
 
 5,776 
 5.50 
 
 722 
 1.52 
 
 9 405 
 1.85 
 
 3,771 
 3.59 
 
 419 
 .88 
 
 10 171 
 
 .78 
 
 1,710 
 1.62 
 
 171 
 .36 
 
 Total 21,801 
 100.00 
 
 105.003 
 100.00 
 
 47,274 
 99.95 
 
c 
 a 
 
 O 
 5 
 
 2 
 O 
 
 O 
 
 m 
 
 o 
 
 <r 
 
MIS: MULTI-CAMPUS ENVIRONMENT 187 
 
 other libraries. Figure 8 compares one SUNY library (Cortland) with the 
 other nine in the analysis. 
 
 The last obtained report is a simple list of OCLC number of titles 
 analyzed, with the OCLC holdings library code arranged alongside the 
 number. This list is used as a crosscheck against the total results, but is also 
 indicative of the ability of the system programs to produce bibliographies 
 by library/libraries, by class, or by discipline. This facility has already been 
 used to produce a bibliography of holdings in African and Afro- American 
 studies for the chancellor's task force. 
 
 Overlap Studies-Derived Results 
 
 The overlap relationships can be described by five curves of highly 
 unique, scarce, moderate, common, and ubiquitous distributions (see figs. 
 9 through 13). The distribution of highly unique material in figure 9 
 indicates a high proportion of unique material (usually 40 to 75 percent). 
 Presumably this indicates a strong or specialized collection. The distribu- 
 tion of scarce material in figure 10 indicates a smaller proportion of unique 
 items; nevertheless a large proportion of the collection is composed of 
 material held by three or fewer libraries. Moderate distribution may be 
 characterized by curves of several shapes (two possibilities are shown in 
 figure 11). It indicates that the largest proportion of the collection is 
 composed of materials held by from three to seven institutions. The distri- 
 bution of common material may be characterized by several shapes (two 
 possibilities are shown in figure 12). It indicates that approximately one- 
 third of the materials are held by five to seven institutions and less than 10 
 percent are held by eight to ten institutions. The distribution of ubiquitous 
 material may be characterized by curves of several shapes (two possibilities 
 are shown in figure 13). It indicates that at least 10 percent of the material is 
 held by eight to ten institutions. 
 
 One interesting area of study was the review of the uniqueness curves 
 by the HEGIS classes. Table 4 reports the distribution among the SUNY 
 libraries. Note that the lowest figure for uniqueness is psychology (47 
 percent), and that area studies (49 percent) is the only other figure below 50 
 percent. Note also that the number of titles at the bottom of column two, 
 47,274, is equal to the number of "actual titles" in the last section of the 
 institution overlap study in table 3. 
 
 A review of the "scarce" titles by number and percent for each SUNY 
 institution by HEGIS class is represented in table 5. These data are helpful 
 in program and acquisitions review. 
 
188 GLYN T. EVANS b ALBERT BEILBY 
 
 INSTITUTION YCM CORTLAND 
 SUBCLASS: 0305 African Studies 
 Total Held in Common With: 
 XBM = 80 
 XFM = 33 
 YBM = 60 
 YGM = 52 
 YPM = 44 
 YOM = 67 
 ZBM = 41 
 ZLM = 78 
 ZPM = 60 
 Total Unique = 15 
 Fig. 8. Volumes Cortland Library Holds in Common with Other SUNY Libraries 
 
 Conclusions 
 
 Work to Date 
 
 A model has been established to describe the library acquisition/reten- 
 tion process, and to support and inform that decision through the manipu- 
 lation of machine-readable data derived as a byproduct of library, 
 publishing and network operations. Where data are available, program- 
 ming has been completed to perform disciplinary and overlap analyses on 
 library holdings as recorded on OCLC tapes. Conversion tables from LC to 
 HEGIS have been established. 
 
 A set of data from SUNY campuses have been analyzed through the 
 programs and the results subjected to further review. These subsequent 
 studies to achieve "derived" results were performed by SPSS and manual 
 
1 23456789 10 
 Fig. 9. Graph of Distribution of Highly Unique Material 
 
 40 
 
 123456789 10 
 Fig. 10. Graph of Distribution of Scarce Material 
 
30 
 
 20 
 
 123456789 10 
 
 Fig. 11. Graph of Moderate Distribution 
 
 30 
 
 20 
 
 10 
 
 1 23456789 10 
 Fig. 12. Graph of Distribution of Common Material 
 
MIS: MULTI-CAMPUS ENVIRONMENT 
 
 191 
 
 30 
 
 20 
 
 10 
 
 1 2345 6789 10 
 Fig. 13. Graph of Distribution of Ubiquitous Material 
 
 analysis, but they are amenable to DSS software when that becomes 
 available. 
 
 Study of the results lead to some interesting conclusions. First, there is 
 a very high degree of consonance in the macrodecisions made by campuses 
 as they select materials. High correlations between discipline demands and 
 acquisition decisions are observed within each campus. Some disciplines 
 are described as being immanent, that is, they have an inherent relation- 
 ship between program and acquisition patterns. We do not yet understand 
 why these phenomena occur. 
 
 On the other hand, at the microdecision (i.e., the decision whether or 
 not to acquire a specific title), remarkable diversity is shown even where 
 programs are apparently similar among campuses. Only two disciplines 
 which have a less than 50 percent uniqueness among ten campuses were 
 found. These results point to four further conclusions: 
 
 1. It seems unlikely that an acquisition formula can be defined. It does 
 seem probable that given the data available in this report, plus the in- 
 corporation of publishing and price data, a statement which describes 
 the campus academic mission bibliographically can be produced and 
 fine tuned over a period of time and in response to changing circum- 
 stance, and provide a firm justification for acquisition budgets. 
 
 2. The access system among campuses is essential. All must belong to the 
 same bibliographic network, maintain their database, use the ILL 
 message network, and the same document delivery system. The diversity 
 among the campuses is the greatest bibliographic asset owned by the 
 
192 GLYN T. EVANS to ALBERT BEILBY 
 
 university. Easy online access and rapid twenty-four hour delivery of 
 documents is the best possible response to fiscal crises. This is, 
 of course, true of any group of libraries which can display the same 
 characteristics as the SUNY libraries. 
 
 3. The bibliographic diversity is a direct result of the subtle diversities of 
 the academic programs at each campus, and a reflection of the book 
 selection processes at each campus. The bibliographic description of the 
 academic mission is a valuable contribution to the academic planning 
 of each campus, and the university as a whole. 
 
 4. Although the results reported are based on current acquisitions, the 
 university libraries at SUNY are engaged in retrospective conversion 
 projects. Access to the whole of the bibliographic resource is the natural 
 concomitant of the strategy proposed in number two above. 
 
 Planned Developments and Extensions 
 
 First, the studies reported in this study must be repeated at a later date 
 to compare the results. The same criteria will be used in the repeat studies. 
 Second, the analysis service is now being offered by SUNY/OCLC as a 
 service to groups of libraries in the state and the country. A number of such 
 analyses have been performed. Their results so far confirm the results 
 reported in this paper. As more results are obtained, SUNY will accumu- 
 late them to build a broader picture and perhaps act as a clearinghouse for 
 such studies. Third, Bowker has now announced the availability of its 
 ABPR and BIP files for such projects as the SUNY project. They must be 
 incorporated in the model. Fourth, the serials database in OCLC is grow- 
 ing rapidly. They must be incorporated in the study. Some preliminary 
 studies are promising. Fifth, DSS software must be incorporated to extend 
 the availability and utility of the derived data. Sixth, use data, primarily 
 from OCLC it is anticipated, must be incorporated in the program. Sev- 
 enth, the U.S. Office of Education has proposed a new and different set of 
 HEGIS codes. Programs must be revised to accept these codes if they are to 
 be accepted by the academic community. We do not have the cost, a time 
 frame, or any sense of the ultimate improvement in the cost efficiency of 
 this step, but if the data produced by the academic institution changes, we 
 have no option. Eighth, ways must found to incorporate the data into 
 academic program planning within institutions. Ninth, the 
 SUNY/OCLC Office of Library Services will seek to extend the utility of 
 the analyses service and acceptance of the reports. 
 
 Prognosis 
 
 Progress on this project has been very slow. It has been difficult to 
 attract support, increasingly so as the fiscal situation has deteriorated over 
 
TABLE 4 
 
 DISTRIBUTION BY HOLDING FREQUENCY OF 
 
 1977-78 TITLES HELD BY TEN SUNY COLLEGES 
 
 "IN Two DIGIT HEGIS CLASSES" 
 
 HEGIS 
 
 Class 
 
 No. of 
 Titles 
 
 Unique 
 
 Holding Frequency 
 2 or 3 or 
 Less Less 5+ 
 
 8-10 
 
 22 Social Sciences 
 
 17,499 
 
 52% 
 
 70% 
 
 80% 
 
 14% 
 
 3.6% 
 
 39 Education 
 
 10,324 
 
 54% 
 
 74% 
 
 84% 
 
 11% 
 
 2.5% 
 
 15 Letters 
 
 8,885 
 
 52% 
 
 69% 
 
 78% 
 
 15% 
 
 3.3% 
 
 21 Public Affairs 
 
 7,172 
 
 54% 
 
 74% 
 
 84% 
 
 11% 
 
 2.7% 
 
 05 Business Management 
 
 4,316 
 
 57% 
 
 75% 
 
 84% 
 
 10% 
 
 2.3% 
 
 09 Engineering 
 
 4,021 
 
 62% 
 
 81% 
 
 90% 
 
 6% 
 
 0.8% 
 
 04 Biological Science 
 
 3,893 
 
 57% 
 
 75% 
 
 85% 
 
 9% 
 
 1.6% 
 
 24 Forestry 
 
 3,474 
 
 61% 
 
 79% 
 
 87% 
 
 8% 
 
 1.5% 
 
 10 Fine fc Applied Arts 
 
 3,398 
 
 50% 
 
 69% 
 
 78% 
 
 15% 
 
 3.5% 
 
 11 Foreign Language 
 
 3,044 
 
 57% 
 
 72% 
 
 81% 
 
 14% 
 
 2.5% 
 
 20 Psychology 
 
 2,716 
 
 47% 
 
 64% 
 
 76% 
 
 16% 
 
 4.7% 
 
 49 Interdisciplinary 
 
 2,599 
 
 65% 
 
 80% 
 
 87% 
 
 8% 
 
 1.2% 
 
 13 Home Economics 
 
 2,502 
 
 52% 
 
 71% 
 
 80% 
 
 13% 
 
 4.0% 
 
 19 Physical Science 
 
 2,329 
 
 57% 
 
 76% 
 
 85% 
 
 9% 
 
 .4% 
 
 12 Health Professions 
 
 ,980 
 
 63% 
 
 80% 
 
 89% 
 
 7% 
 
 .5% 
 
 14 Law 
 
 ,862 
 
 75% 
 
 88% 
 
 92% 
 
 5% 
 
 .4% 
 
 01 Agriculture 
 
 ,628 
 
 64% 
 
 81% 
 
 88% 
 
 6% 
 
 .0% 
 
 02 Architecture 
 
 ,600 
 
 65% 
 
 80% 
 
 88% 
 
 8% 
 
 .0% 
 
 06 Communications 
 
 ,329 
 
 50% 
 
 66% 
 
 77% 
 
 17% 
 
 3.3% 
 
 17 Mathematics 
 
 ,257 
 
 51% 
 
 77% 
 
 88% 
 
 4% 
 
 1.4% 
 
 03 Area Studies 
 
 ,249 
 
 49% 
 
 66% 
 
 76% 
 
 18% 
 
 3.2% 
 
 16 Library Science 
 
 ,116 
 
 56% 
 
 73% 
 
 85% 
 
 7% 
 
 1.0% 
 
 18 Military Science 
 
 799 
 
 52% 
 
 70% 
 
 78% 
 
 15% 
 
 2.7% 
 
 07 Computer Info. Sci. 
 
 563 
 
 62% 
 
 80% 
 
 91% 
 
 4% 
 
 1.4% 
 
 Mean (n = 24) 
 
 
 57% 
 
 75% 
 
 84% 
 
 10% 
 
 2.2% 
 
 Total Acquisitions 
 
 47,274* 
 
 54% 
 
 72% 
 
 81% 
 
 12% 
 
 2.8% 
 
 Note: Because this table is the result of a component analysis, in which one book may be 
 assigned to more than one discipline, 47,274 is the number of titles assigned to disciplines and 
 is not the sum of all assignments. 
 
ij 
 
 u 
 
 C/5 
 
 HH 
 
 O 
 - 
 
 ffi 
 
 W VS 
 
 ft. W 
 
 8 5 
 
 J Q 
 r 1 "* ^ 
 
 w h z 
 J it. S 
 
 PQ O a- 
 
 < a 3 
 
 H w 
 
 BQ 00 
 
 CA1 
 
 - 
 O 
 H 
 
 U 
 U 
 
 U 
 OH 
 
 a? o 
 
 
 
 
 10 ** "2 
 
 
 
 T s^S^ 6*-" 6^2 
 - S oo 
 
 5s 01 
 
 
 
 3 
 I I 
 
 5 
 
 . 
 
 2g S5 1 2J^ 85 Sg 5 
 
 - 
 

 
 
 </> 
 
 
 j 
 
 ? - ! o" - - <! 
 
 % 
 
 
 Vj 
 
 TT CM f) ~ OO **" OO Tp >f) CM or) ~ CM 
 
 '3 ; 
 
 
 1 
 
 
 2 a 
 
 
 
 
 g S 
 
 
 M 
 
 
 n a 
 
 
 
 
 5 go 3 $S gcM $8 
 
 ^ r 
 
 
 | 
 
 ^if, ^CM ^.oo ^- <o^ .f><* ^ -r - 
 
 ^ 5o 
 
 
 
 
 o - 
 
 
 
 
 SI 
 
 
 | 
 
 on fcfj 00 ^| O {-^ CTi TJ- on gj ^D ^ oo ^ OO 
 to *"* ^* ^ ^0 ^ <> ^ r** ^* o "^ ^* ^" ^o " 
 
 s - 
 
 
 e 
 
 
 QO 
 
 
 O 
 
 
 2 oo 
 
 
 
 
 .C *O 
 
 
 1 
 
 1 Ss s II S^ S a IP 
 
 81 
 
 
 
 
 
 s 
 
 
 
 
 fl3 
 
 
 1 
 
 . - ^CM ,,00 ^^ ,,00 g.00 
 
 j= -5 
 -1 
 
 
 1 
 
 kTjOn c^J^* C^l^ CM *O^^ OO^^ CM^^ kO 
 
 S ^ 
 
 
 3 
 
 
 S ro 
 
 "^ 
 
 a. 
 
 
 "u 6? 
 
 c 
 
 
 
 
 S J5 
 
 
 S> 
 
 So? |r>2 SS OCM $S ^^? ^S oo*S 
 
 O 
 
 c 
 
 c 
 
 3^ ^.OO ^^ 00^* t> &t kfj^ Tr^ (COO 
 
 J3 
 
 o 
 
 V 
 
 
 E [ i 
 
 
 
 
 U 
 
 ^^ 
 
 
 
 
 1 
 
 ^ 
 
 
 ; 
 C J 
 
 IT) 
 
 w 
 
 | 
 O 
 
 (O ^ on ^* ^* ^* ^i ^0 25 "^ <O ^ oo ^ oo 
 
 C/5 
 
 aO 
 
 
 
 
 
 "o s 
 
 H 
 
 "5" 
 OQ 
 
 *ft C^ . OJ ^*^o CM^ ^ o< ^ O^ 00 
 
 ts 
 
 
 
 
 l; ^ 
 
 
 2 
 
 
 *j 
 
 
 
 ftp *n hp ^rj ^ oo ftp oo &P oj 6? y? 6^ oo &P 
 
 i -^ 
 
 
 .2 
 
 ^* *2 on ^0 ^* ^ ^* ^ v ^" ^ iTj ^> 0^ CM 
 
 4j ai 
 
 
 8 
 
 ,^00 ^kO ^^" ^'^ ^*^" ^"^ on"" 10"" 
 
 * "3 
 
 
 
 
 
 u '3 
 
 
 
 
 6 "o 
 
 
 | 
 
 oo 1 *^ iftCM r^*^ r>* r*^ oo* S^ ooo 
 
 i| 
 
 
 1 
 
 
 | 
 
 
 
 
 e 
 
 ^2 
 
 
 
 ^ 1 
 
 2l 
 
 / ^ 
 
 
 
 n 6C '3 
 
 
 
 
 I ! Ki1| 
 
 Ji 
 
 
 
 >, O. _ ^ 3 
 
 <r b o < So acMiwgct^c 
 
 CM \ .S 1 OU O'- Og j- 
 
 fcets-c-ge-5 
 
 "^ 
 
 
 
 o . o < *p a q ^ 
 
 u.u,u.J<<U^ 
 
 jj 9J 
 
 
 
 
 O ^ 
 
196 GLYN T. EVANS b ALBERT BEILBY 
 
 the last few years. At the same time, although the service is available, few 
 libraries have taken advantage of it. There is also difficulty in obtaining 
 acceptance and integration of the results into the academic and biblio- 
 graphic decision-making process. It just takes time and patience. 
 
 It is, however, inevitable that, because of their fitness and because of 
 the increased pressures and complexities of decision -making, automated 
 library modeling systems, supported by the analysis of library and other 
 operational data, will gain slow, reluctant acceptance by administrators, 
 budget officials and librarians. 
 
 REFERENCES 
 
 1. "The Shorter Catechism of Stafford Beer." Datamation 28(Feb. 1982): 148. 
 
 2. Spaulding, F.H., and Stanton, R.O. "Computer-aided Selection in a Library 
 Network." JAS1S 27(Sepl./Ocl. 1976):269-80. 
 
 3. Keen, Peter G.W. "Information Systems and Organizational Change." 
 Communications of the Association of Computing Machinery 24(Jan. 1981):24-33. 
 
 4. King, William R. "Using and Evaluating Administrative Decision Support 
 Systems." New Directions for Higher Education 35(Sept. 1981 ):64. 
 
 5. Hopkins, David S.P., and Massy, William F. Planning Models for Colleges and 
 Universities. Stanford, Calif.: Stanford University Press, 1981, p. 4. 
 
 6. King, "Using and Evaluating." 
 
 7. Clapp, Verner W., and Jordan, Robert T. "Quantitative Criteria for Adequacy of 
 Academic Library Collections." College b Research Libraries 26(Sept. 1965):371-80. 
 
 8. New York State Education Department. Guidelines for Assessing the Adequacy of 
 Academic Libraries in New York State. Albany, N.Y.: 1976. 
 
 9. Evans, Glyn T., et al. Development of a Responsive Library Acquisitions Formula: 
 Final Report. Albany, N.Y.: SUNY, Central Administration, 1978. 
 
 10. Beazley, Richard M. Library Statistics of Colleges and Universities: 1979 
 Institutional Data. Washington, D.C.: National Center for Educational Statistics, 1981. 
 
 11. Beilby, Mary H., and Evans, Glyn T. "An Information System for Collection 
 Development in SUNY: A Progress Report." Collection Management 2(Fall 1978):217-28. 
 
 12. Huff, Robert A., and Chandler, Marjorie O. A Taxonomy of Instructional Programs 
 in Higher Education. Washington, D.C.: National Center for Educational Statistics, 1970. 
 
 13. Office of Institutional Research and Analytic Study. Statistical Abstracts (Official 
 report produced by the Course and Section Analysis system containing information on 
 instructional workload, faculty, I&DR). Albany, N.Y.: SUNY, OIRAS, 1982. 
 
INDEX 
 
 Accessions, 79-80 
 
 Acquisition formula budgeting, 167 
 
 Acquisition formulas, inability to 
 
 define, 191 
 Adams, Henry, 104 
 Andreasen, Alan R., 97 
 Anthony, Robert N., 28 
 Automation, and empire building. See 
 
 Empire building, and automation 
 Automation, as management tool, 105 
 
 Beer, Stafford, quoted, 166 
 Bell Laboratories, 128-47 
 BELLCALL, 136 
 BELLPAY, 143 
 BELLREL, 143-44 
 BELLSER, 138-39 
 BELLTIP, 132-37 
 
 Bibliographic analysis, and Geac, 85-86 
 Bibliographic cooperative manage- 
 ment, and management information, 
 
 117 
 Bibliographic information, as barrier 
 
 to use of online catalog, 120 
 Bibliographic Retrieval Service, See 
 
 BRS 
 Bibliometrics, and online databases, 
 
 150-52 
 
 Billing information, 83-84, 1 15 
 Book selection, at Bell Laboratories, 
 
 130-36. See also Selection of materials 
 BRS databases, 151-55, 158-59 
 Budget and expenditures, at PPLD, 33, 
 
 37 
 Budgeting, acquisition formula. See 
 
 Acquisition formula budgeting 
 
 Caldwell, Jane, 151 
 
 Cataloging information, at Bell Labor- 
 atories, 136 
 
 Centralization, and management infor- 
 mation systems, 63-64 
 
 Challenged material, 55 
 
 Cheney, Paul H. 28 
 
 Cherns, A.B., 105 
 
 Choice, in decision making, 13-15 
 
 Churchman, C. West, quoted, 15-16 
 
 Circulation information, 74-82 
 
 Clapp- Jordan formula, 167 
 
 Class reserves, 80-81 
 
 Clerical tasks, automation of, 27 
 
 Cocitation analysis, 156 
 
 Collection development, and online 
 
 databases, 151-54 
 Collection management, and Geac, 85- 
 
 86 
 Command language, as barrier to use of 
 
 online catalog, 120 
 Communication, breakdown in, in 
 
 organizations, 97-100 
 Compliments and complaints, at 
 
 PPLD, 33 
 
 Component analysis, 175-80 
 Computerization, as barrier to use of 
 
 online catalogs, 119 
 Control process, cybernetic, in systems, 
 
 18-20 
 
 Cost, of data. See Data, cost of 
 Critical incidents, at PPLD, 50 
 Cultural deficiencies, as cause for non- 
 use of information, 102-03 
 Current information alerting service, 
 
 139-43 
 Cybernetics, 18-20 
 
 Data: collection of, 3-4, 114; cost of, 11; 
 definition of, 10; historical nature of, 
 114; for management information 
 system, 21, 171; sifting of, 112 
 
 Data Entry Report (Geac), 80 
 
 Data rejection, reasons for, 5 
 
 Data system, and congruency with 
 organization, 20 
 
 Database, definition of, 148-49 
 
 Database searching, 149 
 
 Decentralization, and management 
 information systems, 63-64 
 
 Decision-making process, levels of, 13- 
 16 
 
 Decision support models, 22 
 
 Decision support system, 28-58, 166; 
 functions of, 29-30 
 
 Deficiencies, educational. See Educa- 
 tional deficiencies 
 
 Design, in decision making, 13-15 
 
 Dickson, Gary W., 28 
 
 Digital Equipment Corp., 43 
 
 197 
 
198 
 
 INDEX 
 
 Dowlin, Ken, 29 
 
 Drucker, Peter, quoted, 1 1 1 
 
 Educational deficiencies, as cause of 
 nonuse of information, 100-01 
 
 Ellingson, Celia, 151, 152 
 
 Empire building, and automation, 103- 
 04 
 
 Employee satisfaction, and manage- 
 ment information systems. See Man- 
 agement information systems, and 
 employee satisfaction. 
 
 Evaluation: of collection and online 
 databases, 152; of management infor- 
 mation systems, 21 
 
 Exchange process, 93-100 
 
 Expenditures, flow through. See Flow 
 through expenditures 
 
 Failure, reasons for, of management 
 support systems. See Management 
 support systems, failure of 
 
 Feedback control mechanisms, in sys- 
 tems approach, 19-20 
 
 Financial information, in Geac, 82-84 
 
 Financial records, of individual user, 84 
 
 Fines (Geac), 84-85 
 
 Firnberg, David, 101 
 
 Flow through expenditures, at PPLD, 
 43 
 
 Formulas, acquisition. See Acquisition 
 formulas 
 
 Geac, 72-91 
 
 General ledger program, 160-63 
 GLUG (Geac programming language), 
 76 
 
 Hawkins, Donald, 150, 151 
 
 Hedstrom, Lori, 152 
 
 HEGIS data, 116, 171; and relation to 
 LC class no., 172 
 
 Higher Education General Informa- 
 tion Survey. See HEGIS 
 
 Hildreth, Charles R., 119 
 
 Hold status, for books, 77-79 
 
 Horn, Roger, 96 
 
 Information: definition of, 10; instan- 
 taneous access to, 113-14; integration 
 of, 124; nonuse of, cause of, 100-02; 
 and organizational control, 106; 
 value of, 1 1 
 
 Information and Reference Report, at 
 
 PPLD, 50 
 
 Information budgets, 8 
 Information gathering, and automated 
 
 systems, 106 
 
 Information need, 113; and time con- 
 straints, 120-22 
 Information overload, 101-02 
 Input-process-output model, 12 
 Integration, of information. See Infor- 
 mation, integration of 
 Intelligence, in decision making, 13-15 
 Interagency Activity Report (Univ. of 
 
 Arizona), 74-75 
 
 Interlibrary Loan Report, at PPLD, 55 
 Internal orientation, 96-97 
 Inventory count, at PPLD, 33, 41 
 Inventory use ratio, at PPLD, 30, 35, 44, 
 
 45, 46 
 Inventory use ratio exceptions, at 
 
 PPLD, 35, 42 
 Item information, 5 
 
 Joseph, Earl, 125 
 
 Kaske, Neal, 118 
 Keen, Peter 166 
 Known-item searches, 121 
 
 LC class number, and HEGIS file, 172 
 Libraries, systems view of, 16-20 
 Library data, and lack of link with aca- 
 demic administrative data, 168 
 Library operations, models of, 7 
 Lost items. See Missing items 
 Luddite response to automated infor- 
 mation systems, 102 
 
 Macrodecisions, and selection of mate- 
 See Selection of materials, and 
 macrodecisions 
 
 Management control, 28 
 
 Management information: and deci- 
 sion situation, 10; and individual 
 manager, 11; from online circulation 
 system, 143-45; and organizational 
 context, 11; users of, 116-17; uses of, 
 169 
 
 Management information systems: 
 acceptance of, variables affecting, 60- 
 62; and alteration of organizational 
 processes, 65-66; analysis and design 
 
INDEX 
 
 199 
 
 stage of, 24; budgets for, 8; centraliza- 
 tion v. decentralization, 63-64; and 
 changes in control, 64-65; and com- 
 puters, 6-7; v. decision support sys- 
 tems, 166; desiderata for, at Univ. of 
 Arizona, 73; design of, 29; and effect 
 on management, 67-70; and 
 employee satisfaction, 69-70; evalu- 
 tion of, see Evaluation, of manage- 
 ment information systems; impact of, 
 on organizational processes, 66; mo- 
 tivation to use, 105-06; organiza- 
 tional aspects of, 59-70; planning for, 
 24; and professionalization, 64; pur- 
 poses of, 111-12, 169; stratification of, 
 4; at SUNY, reasons for inception, 
 167-68 
 
 Management Review and Analysis Pro- 
 gram. See MRAP 
 
 Management support system, failure of, 
 22 
 
 Manager, and management informa- 
 tion, 116 
 
 Managerial action, 10 
 
 Marketing, problems of. See Exchange 
 process 
 
 Markey, Karen, 118 
 
 Materials encumbrances, 50, 52-53 
 
 Mathematics, fear of, 7 
 
 MERCURY, 139-43 
 
 Microdecision, and selection of mate- 
 rials. See Selection of materials, and 
 microdecision 
 
 Middle manager, changes in work of, 
 with management information sys- 
 tems, 67 
 
 Miller, George, 8 
 
 Missing items, 75-77 
 
 Model, for management information 
 system, 169-70 
 
 Modeling, 166-67 
 
 Models, for managerial support, 22 
 
 Mooers, Calvin, 2 
 
 Mooers's Law, 2, 5 
 
 Motivation, to use management infor- 
 mation systems. See Management 
 information systems, motivation to 
 use 
 
 MRAP, 68 
 
 National Center for Higher Education 
 
 Management, 31 
 Numerical information, and difficulty 
 
 of use, 100-01 
 
 OCLC Interlibrary Loan Subsystem, 
 115 
 
 Online Catalog Project, 119 
 
 Online catalog use count, at PPLD, 43, 
 49 
 
 Online catalogs, barriers to use of, 1 19- 
 20 
 
 Online databases: and bibliometrics. 
 See Bibliometrics, and online data- 
 bases; and collection development. 
 See Collection development, and 
 online databases 
 
 OPACS. See Online Public Access 
 Catalog Systems, 118 
 
 Operational control, 28 
 
 Organization: input/output processes 
 of, 12; maturity of, 61; psychological 
 climate of, 61; as source of manage- 
 ment information, 20; subunit of, 12; 
 systems view of, 11-20 
 
 Organizational control, and informa- 
 tion. See Information, and organiza- 
 tional control 
 
 Organizational processes: at general 
 personnel level, 66; at individual 
 level, 65-66 
 
 Organizational resources, budgeting of, 
 61 
 
 Overdue items, 85 
 
 Overlap analysis, 175-76, 180-88 
 
 Overlap studies, of databases, 154-55 
 
 Owens, Tom, 73, 74 
 
 Parkinson's Law, 7 
 
 Patron information: at PPLD, 55, 57; 
 
 with Geac, 87 
 
 Patron tenure, at PPLD, 43, 48 
 Patron use, at PPLD, 43, 47 
 Patron use ratio, 30 
 Perceptual survey, at PPLD, 35, 42 
 Performance gaps, 94-96 
 Periodicals use, at PPLD, 55, 56 
 Pikes Peak Library District, 27-58; 
 
 organizational structure of, 36; 
 
 reports produced by, 33-58; survey of, 
 
 31-33 
 
200 
 
 INDEX 
 
 Planning, for management informa- 
 tion systems. See Management infor- 
 mation systems, planning for 
 
 Power: and management information 
 systems, 124; and special knowledge/ 
 information, 107-08 
 
 PPLD. See- Pikes Peak Library District 
 
 Priesthood, of automation, 103-05 
 
 Printouts, 7 
 
 Privacy, problems of, 124 
 
 Processing, efficiency of, 112 
 
 Project status reports, at PPLD, 35, 55 
 
 Reference questions, and online data- 
 bases, 149-50 
 
 Regional network managers, and man- 
 agement information, 117 
 
 Rejection, of data. See Data rejection 
 
 Research, and management informa- 
 tion systems, 118-21 
 
 Research interests, of patrons, and 
 online databases, 155 
 
 Revenues, at PPLD, 33, 38 
 
 Runyon, Robert S., 93 
 
 "Russell's inference," 2 
 
 Scattering, 151 
 
 Search log, information for, 157-58 
 
 Security reports, at PPLD, 43 
 
 Selection of materials: and macrodeci- 
 sion, 191; and microdecision, 191 
 
 Senior management, and managment 
 information, 1 17 
 
 Sensory deficiencies, as cause for non- 
 use of information, 101-02 
 
 Serials control, at Bell Laboratories, 
 138-39 
 
 Service data, lack of, 168 
 
 Service indicators, at PPLD, 33, 39, 40 
 
 Shank's syndrome, 2 
 
 Strategic contingencies model, 107-08 
 
 Strategic planning, 28 
 
 Stress, in organizations, 94-96 
 
 Subject access project (OCLC), 118 
 
 Suboptimization, 112 
 
 SUNY/OCLC Network, 165-66 
 
 System development process, 23-25 
 
 Systems approach, of C.W. Church- 
 man, 15-18 
 
 Systems management report (Geac), 87- 
 89 
 
 Terminal Requirements Project 
 
 (OCLC), 118 
 Tolle, John, 118 
 Transaction analysis, 122-23 
 TSTA Report, 81 
 
 University of Arizona, library manage- 
 ment information system of, 72-91 
 User cards, expired, 77 
 
 Value, of information. See Informa- 
 tion, Value of 
 Visual illiteracy, 7