Library Services in Theory and Context

Chapter 10: Demand

Definition

Various sorts of needs and wants with respect to library service have been discussed in the literature of librarianship. These include:

1. needs that are not recognized as needs—or not recognized as needs for which library services would be useful;
2. needs that are recognized but, nevertheless, no action is taken to use a library service in relation to them;
3. wants (in the sense of desires) to use a library service, whether or not such use is sensible in practice;
4. unsuccessful attempts to use a library service, as when a particular book is sought but is not found;
5. satisfied demand, in the sense that the library was used in a way that was satisfactory to the user.

A variety of terms have been coined in relation to these various sorts of needs and wants. ¹ However, we will deal directly only with the last two; both of which represent actual use of a library service. This constitutes the expressed demand (in an economic sense) with which the supplier of library service is dealing. As such, demand is a major determinant of what happens in library services.

However, merely to define demand does not provide much understanding. If we are to have any serious understanding of the structure and functioning of library services, we need to ask: What determines the demand for library services? We also need to know something of the dynamics of the demand for library services.

In exploring the determinants of demand, we take what is basically an economic approach except that we prefer to view "economic man" as being "cybernetic person." Whatever phrase is used, we mean a process with the following characteristics: individuals have desires that, in their own perceptions, could be satisfied by using a library—or, at least, the probability is high enough to warrant a try. Individuals weigh the perceived probable "price" of using the library service against the perceived probable benefit of doing so. Library use is likely to follow if the relative price is low enough. "Relative" not only means in comparison with the expected benefits of use but also in comparison with alternatives to that particular library service.

In exploring this theory of demand we will consider two of the paradoxes involved:

1. An economic theory based on price appears inconsistent with the fact that library services are ordinarily free.
2. If library services are provided for free, then libraries would appear to lack the responsiveness associated with market forces. Hence the suggestion that libraries ought to charge in order to become responsive. ² Then, how is it that libraries exhibit considerable stability and powers of survival, which are characteristics usually associated with systems that are highly responsive to changing circumstances?

The resolution to both paradoxes comes with a reconsideration of the concept of price.

The real price

Economics texts vary in their treatment of the definition of price. Sometimes, price is defined as the monetary exchange value of commodities and services. Sometimes, that definition is implicitly assumed. At other times, a more general definition of price is given and, subsequently, the notion of money as a convenient mode of expression of the price is added as an extension:

Price, measure of the value of a commodity that expresses its worth in exchange for other goods and services. Because it is more convenient to express the relative values of all goods in terms of a common unit of measure. ... In modern economies, the national currency serves this function. ... ³

Adam Smith was emphatic on price being the sum of the disadvantages accruing to the purchaser: "The real price of everything, what everything really costs to the man who wants to acquire it, is the toil and trouble of acquiring it." ⁴

If the "real price" is distinguished from the monetary expression of price, then we can reappraise both the nature of price and the applicability of price mechanisms—and, in particular, any nonmonetary expressions of price—especially in relation to public services. For convenience, the aspects of price will be considered under four headings: time, monetary cost, effort, and discomfort. We shall continue to use the library service as our example, but we do so in the belief that the discussion is of much wider applicability.

Time: delay as an expression of price

Other things being equal, sooner is better than later for anything one wants to have. There is a distinction to be made between waiting for something (during which time one might be able to use the time doing other things) and being obliged to spend much time doing or making it. In both cases, however, an opportunity cost is implicit. Waiting for a book to be made available (e.g., acquired, processed, fetched from the stacks, recalled from another user, obtained on interlibrary loan) is an unwelcome delay. There may be no doubt that a library service in which these delays are short is better than one in which they are long, other things being equal. Best of all is the library service in which books are immediately available, with the necessary tasks of preparation already performed by the time the demand occurs. One prefers not to have to expend the time or incur the delay to get the service. ⁵ (For a detailed treatment of time as a cost in the use of library services see Van House.) ⁶

Money as an expression of price

The use of money as an expression of price needs no explanation. It is included here for completeness. One prefers to minimize (ideally to avoid) the expenditure of money to obtain service. Library services are generally free to those who use them. The exceptions (such as photocopying, online literature searching) tend to be reluctantly imposed, subsidized to keep costs down, and made free when possible. Fines for the late return of borrowed materials are better seen as a reluctantly imposed control device intended to sustain the level of service. If books do not come back, they are not available for others. Fines for overdue books are not a price in the normal sense, although they could be viewed as the exchange value for lateness. There is a sizable literature on rationales and techniques for pricing library services but that is not directly relevant here. ⁷

Effort and discomfort: inconvenience as an expression of price

Both time and money vary in how they are valued by individuals according to the circumstances. Under these two factors (effort and discomfort), we include a number of considerations that are even less easy to handle than time and money. However, we are primarily interested in the "real price" rather than whether or not its, expressions lend themselves to convenient quantitative analysis. Like elephants, inconveniences are easier to recognize than to describe. It is a nuisance to have to travel in order to obtain service. An overheated (or underheated), noisy reading room with a distracting decor makes use of a library service less pleasant. These factors can be cultural and interpersonal as well as physiological. One would prefer not to have to put up with such inconvenience in order to use a library service. If the inconvenience were great enough one would be deterred from using that library. Another library service which was less inconvenient, but otherwise comparable would be preferable.

These aspects of the "real price" invite several comments. It is clear that there can be nonmonetary elements of price in "the toil and trouble of acquiring it." Indeed, the monetary element may be minor or nonexistent. ⁸

The question arises as to whether some of these aspects of price ought to be considered as part of the price at all. For example, delay could be viewed as much an attribute of the service or commodity as of the price. From an analytical point of view, there is scope for flexibility here. All of the elements of a cost-benefit relationship are part of the same equation, and it may not matter how one chooses to arrange them. The argument that monetary price is an incomplete expression of price could be countered by the response that, by definition, monetary price is being isolated on one side of the equation and that any aspects not expressed by monetary price can be regarded as attributes of the service or commodity (e.g., a slow service, an inconvenient service). Theoretically, there is a difference between using monetary price as a convenient but incomplete representation of exchange value (in which case something has been lost) and asserting that monetary price is what happens to have been isolated on one side of the equation (in which case nothing has been lost).

Each aspect has the same sort of effect in terms of a price mechanism. As the time needed to obtain the service increases, the customer is more likely to give up. This may involve switching to an alternative source of supply or deciding to forego satisfying the need altogether. The elasticity of demand (the extent to which demand changes if the price changes) is well established with respect to monetary price. There are also limits to the amount of inconvenience people are willing to suffer to satisfy their needs. As with monetary price, the point at which people decide that the amount of inconvenience is too much can be expected to vary from one individual to another and from one situation to another. However, there appears to be no reason to doubt that elasticity of demand exists in library use with respect to the nonmonetary aspects of price as well as with monetary prices. ⁹ The more trouble it is to use a given library, the less one is inclined to use it. Given a choice of sources of service with different degrees of inconvenience, people will tend to choose rationally between them.

These different aspects of price are more or less interchangeable. People may be willing to pay to avoid delay. Others may be willing to save money by waiting, if, in their personal value system, money is scarce relative to time. Similarly, effort and/or discomfort may, for any given situation, be more (or less) preferable to delay or a monetary price. Every competent bureaucrat understands that the choice of process can affect the outcome and that a price mechanism can be invoked to discourage demand even without a monetary price: increased delays, more forms to complete, inconveniently located office, restricted hours of service, redirection from one office to another, instructions that take time to understand are but a few examples. Whether intentional or not, these increases in the nonmonetary aspects of price all have a dampening effect on demand just as the imposition of (or an increase in) monetary price does. If there were complete interchangeability between the aspects of price and if there were perfect understanding of what the exchanges were, then all aspects of price could, very conveniently, be related to money. Such a degree of interchangeability appears to occur only in economic theory.

With this general background, let us return to library services and consider what could happen if the demand for a particular text were to increase. ¹⁰ Sooner or later, all copies of that title in the library will be in use and demands for copies will be going unsatisfied. From the library's side there are at least four possible responses:

1. Acquire additional copies. Assuming no increase in the library's budget, this implies some internal reallocation of resources is feasible and acceptable. The price paid by users will include an opportunity forgone elsewhere: a degradation of service in some other respect. (This might well constitute a net improvement in the library service overall, but some disbenefit elsewhere in the library service is ordinarily involved, however slight.)
2. Increased availability could, alternatively, be achieved by faster turnaround. In other words, the librarian could shorten loan periods and/or confine copies to the reading room, so that they will become available for others all the sooner. In this case, instead of imposing a price in terms of a reduction of service elsewhere in the library (as when diverting money to buy extra copies), the service is made less convenient in a different way for the users of that title. Whether they will accept the increased inconvenience and still use the library's copies or decide that they are unwilling to tolerate the reduced convenience, the library has, in fact, responded to regain earlier levels of availability.
3. The library could introduce a monetary charge for preferential treatment. This would be a commercial market response by the library. This move should be expected to reduce demand, and it would bring in some income to improve the service (e.g., by acquiring more copies), and it might make the waiting more tolerable for those who prefer not to pay but realize that they could pay for faster service if they valued their time highly enough relative to their money.
4. A common response is to do nothing. At least, the librarian can decide to adopt no specific response to restore the level of availability. In this, the time aspect of price, the delay, is being increased for the library users.

It is clear that the librarian has more than one possible response. It is also clear that each response improves the service but also increases the real price to the library users, but in substantially different ways. The choice of response by the librarian implies a decision to change the mixture of elements in real price to the library users for using the library.

The response of the user of the library to a change in the real price can be expected to be the same as in any other situation of demand for a commodity or service: the customer can accept the price involved (mainly time and inconvenience in this case), can substitute another source (go to another library, borrow the personal copy of a friend, or buy a copy), can accept a substitute service (a copy of a different title in lieu of the one originally sought), or decide to give up, leaving the need unsatisfied. Which decision any given individual will make will depend on the circumstances: the alternatives available; the importance, urgency, and specificity of the need; and the various values held by the individual with respect to time, money, and inconvenience.

The customers' decisions determine the level of demand for the title. If the increase in demand is dampened enough, the balance of demand and supply will return to where it was originally. In this discussion, we have chosen an aspect of the logistics of library service to provide an example because it seems to illustrate the dynamics of the situation rather well. However, we believe that the same sort of mechanism applies generally.

The double feedback loop ¹¹

Of fundamental importance to the understanding of the dynamics of library services is the separateness of the responsive behavior of the user and that of the librarian. Let us illustrate this by considering further the example in the previous section of an activity central to library service: somebody looking for a copy of a particular book. Assuming that the library does, in fact, own the desired title and that the user has correctly identified its place on the shelves, ¹² the probability of a copy being available when wanted depends on the interaction of three variables: the pattern of demand; the number of copies owned; and the length of time copies are unavailable when being used.

If the librarian knows that the availability of a particular title is low, then availability can be increased by adding extra copies and/or reducing loan periods. Such changes constitute adaptive responses (feedback) by the librarian to improve an unsatisfactory standard of service.

If the user does not find the book, then the user can respond in either of two sorts of ways: by coming back later, in which case the demand remains; or by taking other steps which will have the effect of reducing the demand on the document. One can switch to another library, buy a personal copy, use a friend's copy, or simply give up. This freedom of choice by the user constitutes a second, independent feedback mechanism. The second set of options result in a lowering of demand and, thereby, increased availability.

This double feedback loop (figure 10.2) is of interest from a cybernetic perspective. These two feedback mechanisms are substantially independent of each other in important ways. ¹³ The librarian's action does not depend

Fig. 10.2 The double feedback loop in library service

on response by the user; and the user's action does not depend on response by the librarian. Further, since library services are normally free, the library's income does not depend directly on the level of demand. Reduced demand, therefore, does not weaken the library as it would a business, where a drop in demand would reduce sales and, therefore, income. Quite the reverse, a reduced demand for a free service reduces the pressure, leaving the existing resources more adequate to cope with the remaining demand. (Similarly, serving increased demand may not be rewarding as it would for a commercial business.)

The dynamics of users' responses to the experience of trying to use library services has hardly been studied and its nature is not yet well known. However, the demand for library services appears to be elastic with respect to a price mechanism based on reliability, delay, convenience, ambience, and, where applicable, money. The adaptive aspects of library demand can be viewed as stabilization by the user with respect to a balance of effort and benefit. If the price of trying to use a library service is perceived by the users as increasing relative to the benefit and/or the alternatives, the frequency of use can be expected to diminish.

Although the librarian can respond to improve availability, such response is not very probable because, with present library techniques, evidence about the unsatisfied searches by users commonly does not reach the librarian in any reliable, useful way.

However, the user always knows whether or not the book was found. The user's feedback mechanism will be in effect, whether or not the librarian responds. In other words, it is the responsiveness of the users, more than the responsiveness of the librarians, which serves to restore stability to the library service. This helps to explain a noteworthy cybernetic aspect of library services: Library services can survive with remarkable stability even in the absence of effective library management. Survive, that is, not excel.

As the structure and dynamics of library service have come to be more explicitly understood, it has become possible to make the library's feedback mechanism more effective. For example, variable loan and duplication policies can relate the length of loan and the purchase of duplicate copies more directly to pressures of demand. Hence, changes in the level of demand can induce an adaptive response by the library. The library service can, thus, become more responsive to demand with respect to the maintenance of standards of service—in this case, the probability of desired documents being available when sought. The librarian's role can then move from a primary level of control (responding directly to needs as perceived) to a secondary level of control (adjusting built-in feedback mechanisms to maintain such standards of service as are deemed affordable). Computer-based systems for recording loans can provide, as a byproduct, the necessary management information.

However, our present concern is not so much with the mechanics of book availability but, rather, with illustrating the double feedback loop&151;the separate identities of supply mechanisms from demand mechanisms in libraries where we can assume that the funding of the library service is not derived from charging the user. (When this assumption ceases to be true, the two feedback loops begin to merge into one as will be discussed in chapter 12.)

In the next section we discuss further the ways in which the demand for library services might be sensitive to various factors.

The sensitivity of demand

Historically, research in librarianship has been largely dominated by studies of retrieval and of the history of the book. Study of the dynamics and sensitivity of demand has been seriously neglected. There have been thousands of "user studies," but these have generally been surveys at particular points in time. Hence, as with a photograph, one may get a clear view of the situation at a given moment, but learn little or nothing of the trends, let alone the dynamics of the situation. This is a serious criticism, since applicable research generally involves trying to answer the question: "What would happen if ... ?" If it is not at all well known how demand would change if the provision of library services were to be altered, then one knows very little about "What would happen if ... ?"

It would be an exaggeration to say that there have been no studies leading to models of the dynamics of users and/or of library services. For example, the phenomena examined in bibliometric studies do reflect patterns of user behavior. ¹⁴ There have also been a few formal model-building studies of library use. ¹⁵ Nor should we ignore the knowledge and "feel" which come with experience of library practice yet may not be reflected in the research literature. What is seriously lacking, however, are models of information-gathering behavior, especially marketing models of consumer preferences in relation to library services.

It is known that accessibility is a dominant factor in information gathering behavior. Perceived convenience influences greatly the choice of source of information.¹⁶ The probability that documents will be immediately available when wanted—itself an index of convenience of accessibility—has also emerged as influential with respect to demand. ¹⁷ The opening of a new more congenial library building is known to increase use of library services. ¹⁸ The geographical location of a library also affects the amount of use made of it. ¹⁹ All of these effects are to be expected. Using a reasonably planned new library building ought to involve less discomfort for the user than an older one; a site that is conveniently located minimizes travel effort; immediate availability saves time and effort.

These aspects of the real price do not lend themselves to direct quantification, but that fact does not make them any less real. For example, the effect of distance has been studied by various investigators, but the problem emerges as considerably less tractable than might appear at first sight. For example, it is not a matter of actual physical distance but rather of traveling time. Further, since more than one errand can be performed in one journey, the effort of making the journey can properly be shared between different goals, e.g., between going to the library and going shopping. Nor are all modes of traveling equally congenial, nor the same mode at different times, since one can be more or less busy and the weather more or less pleasant. These factors all conspire to make quantification difficult.

As with other human perceptions of value, the appraisal of the "price" of distance is not linear. In other words, an increase in distance from 10 meters to 100 meters is much more significant than an increase from 100 meters to 190 meters, let alone an increase from 10,000 meters to 10,090 meters, even though each example constitutes an increment of the same amount: 90 meters. In terms of perceived and actual impact on behavior, changes from 1 meter to 10, from 10 to 100, and from 100 to 1,000—geometric as opposed to linear progression—are more likely to have comparable effects. ²⁰

Even though the dynamics of library use are poorly understood and quantitative analysis of the behavior of individual library users is difficult, some general "laws" or patterns have been discovered in information-gathering behavior. In the next section, three general laws that reflect patterns in the demand for library services will be described.

Three patterns of demand: scattering, decay, and inertia

Quantitative studies of the use of library services have resulted in the discovery of three patterns of usage so frequently that each has been referred to as a "law": scattering, decay, and inertia. Between them they do much to define the "shape" of the use of library services.

Scattering. the variation in demand from title to title

It is obvious that some books and articles are used more than others. For every book in a library, there are thousands that the librarian has not seen fit to acquire because the expected usage is deemed too low. Within every library, there are books that have remained unused for years and others that are in heavy demand.

The literature of a subject can be represented by a comprehensive list of articles on that subject. This list can be regarded as representing the literature needed by library users interested in that subject. Analysis of such a list will indicate which periodicals contain a large number of these articles, and it can be presumed that these periodicals will be more useful to this group of users than periodicals that contain very few or none of the articles.

Citations in articles or books can be analyzed on the assumption that an author will refer only to items useful for an understanding of his topic. The totality of such citations can, therefore, be regarded as an approximation of the useful literature of a subject as defined by the group of articles and books whose citations were analyzed—commonly "the ten leading journals on ..." Periodicals that are relatively heavily cited are likely to be more useful than those that are cited little or not at all.

Actual usage of items as recorded on a library's borrowing records, requisition slips, or other records is relevant in that it reflects the usage made of the stock of a library by its readers. Certainly, the data will lack details of items used inside the library, usage of copies personally owned or borrowed from other libraries, or attempts to use items not found in the library. ²¹ However, attempts to make explicit adjustments for these factors rapidly become unmanageable and the adjustments for the last two are unnecessary if one is concerned with measuring the actual use of a particular library as opposed to measuring the cosmic importance of a given title.

It should be clearly admitted that all three methods of analysis measure different things, but all three types of data can be treated as approximate guides to the variations in usefulness of specific titles to different groups of users, even though all three methods are open to criticism on theoretical grounds. All three, however, are practical techniques and, what is more important, the literature available seems to suggest that all three give quite similar results.

The pioneer of the analysis and description of the variation in demand from title to title was S. C. Bradford, then Librarian of the Science Museum Library in London. He examined the literature of applied geophysics and the literature of lubrication. ²² In each case, he counted the number of references to each periodical title. This indicated the variation from title to title in each subject. He then ranked the titles according to their productivity (the number of references contributed by that periodical to the literature concerned) and created a description of each literature by drawing a graph showing the number of references contributed by the single most productive periodical, the number contributed jointly by the two most productive periodicals, the three most productive periodicals, and so on. Naturally, the additional number of references contributed by successively less productive periodicals became fewer and fewer. However, Bradford noticed that this decreasing productivity followed a recognizable pattern; and, since he had been writing in terms of the "scattering" of a literature over journal titles, this pattern came to be known as "Bradford's Law of Scattering."

The pattern identified by Bradford took on a new significance with the realization that it was, in effect, a law of diminishing returns with respect to the increase in the number of titles in a collection—always assuming that more useful titles are acquired before less useful titles. Seen in this perspective, the pattern that Bradford perceived becomes not so much a statistical curiosity in subject bibliography as a potentially powerful tool in the economic analysis of library provision. ²³

Decay: the variation in demand for a title through time

The demand for library services is characterized by chronological patterns. Most of the research in this area has been concerned with the variation in usage from year to year. It has been found consistently that the annual usage of books declines ("decays") with age.

Although a negative exponential pattern of obsolescence is generally accepted for both monographs and periodicals, rigorous analysis becomes very difficult for reasons that have been stressed in Brookes' lucid treatment of obsolescence ²⁴ and also by Fussler and Simon. ²⁵

There are two views of the "decay" of demand:

• In a diachronous view, one is concerned with the use of a given document in successive years—"through time"; and
• In a synchronous view, one is concerned with the distribution of use made during a given span of time of documents of different ages.

Either way, a similar pattern tends to emerge and, like scattering, can be regarded as a law of diminishing returns with respect to the length of time books are retained. ²⁶

Inertia: the variation in demand with distance

Earlier in this chapter we discussed effort as being a significant part of the price involved in the use of library services. One particular manifestation of the effect of human inertia—or, more elegantly, economy of effort—may be seen in the decline in the use of libraries as the distance to be traveled increases. ²⁷

These three empirically found patterns appear to reflect mechanisms that are not yet well understood. Numerous studies have pointed to the existence of such patterns, but very little systematic attention has been paid to the problem of explaining why they are found.

The rate of decay in the demand for books with time is generally regarded as varying from one field of discourse to another. So also, it would seem, does the rate of scattering. There is some indication that this variation is associated with variations in the "hardness/softness" of different areas of study, the extent to which concepts can be unambiguously defined. After all, if subject areas do, indeed, vary in their "hardness/softness" and if this does reflect some variation in the definability of concepts, then one should expect "hardness/softness" to emerge as an important variable in the study of information. Further, one should also expect to find manifestations of it in empirical patterns of information use. ²⁸

Even though the inner workings of the demand mechanism determining the patterns of the use of library services are not well understood, some of the effects can be observed empirically. Three of these effects can be seen in regularities in the distribution of demand from title to title, in the distribution of demand for titles through time, and in the variation in demand with distance. The dismal-sounding trinity of scattering, decay, and inertia warrant attention because they are manifestations of user behavior and, as such, deserve to be central in the design of library services.

In the next two chapters we move from discussion of the demand for library services to consideration of the provision of library services. How do they come to be provided in the manner and to the extent that they are?

Go to Chapter 11

---

¹ M. B. Line, "Draft Definitions: Information and Library Needs, Wants, Demands and Users," Aslib Proceedings 26, no. 2 (February 1974): 87; N. Roberts, "Draft Definitions: information and Library Needs, Wants, Demand and Users: A Comment," Aslib Proceedings 27, no. 7 (July 1975): 308-13; and T. D. Wilson, "On User Studies and Information Needs," Journal of Documentation 37, no. 11 (March 1981): 3-15.

² R. L. Ackoff et al., The SCATT Report: Designing a National Scientific and Technological Communication System (Philadelphia: University of Pennsylvania Press, 1976), p. 41. Cf. M. Getz, Public Libraries: An Economic View (Baltimore: Johns Hopkins Press, 1980), pp. 162-3.

³ Encyclopedia Britannica. 15th ed. 1977. Vol. VIII, p. 204. s.v. "Price."

⁴ A. Smith, An Inquiry Into the Nature and Causes of the Wealth of Nations (Oxford: Clarendon Press, 1976). Book 1, chapter 5, para. 2.

⁵ Cf. "Waiting time does allocate public services, rationing them, as would money prices, according to the tastes, income and opportunity costs of consumers. Time prices differ from money prices, however, since they appear relatively lower to persons with a lower money value of time. ... Thus when two individuals who value their time unequally wait in the same queue, they face different prices." D. Nicols, E. Smolensky and T. N. Tideman, "Discrimination by Waiting Time in Merit Goods," American Economic Review 61, no. 3 (June 1971): 312-23.

⁶ N. Van House, "Time Allocation Theory of Public Library Use," Library and Information Science Research 5, no. 4 (Winter 1983): 365-84. N. Van House Public Library User Fees: The Use and Finance of Public Libraries (Westport, Conn.: Greenwood Press, 1983). This monograph is a revised version of her doctoral dissertation: N. V . H. DeWath, "Demand for Public Library Services: A Time Allocation and Public Finance Approach to User Fees." Ph.D. dissertation. University of California, Berkeley, 1979. (Ann Arbor, Mich.: University Microfilms, order no. DEM 80-14651.

⁷ For example, C. A. Casper, "Pricing Policies for Library Services," Journal of the American Society for Information Science 30, no. 5 (September 1979): 304-9. See also C. K. Mick, "Cost Analysis of Information Systems and Services," Annual Review of Information Science and Technology 14 (1979): 37-64.

⁸ Y. M. Braunstein, "Costs and Benefits of Library Information: The User Point of View," Library Trends 28, no. I (Summer 1979): 79-87.

⁹ The evidence is slight. Indeed, Casper found grounds for suggesting a lack of elasticity of demand, but that was in rather special circumstances: C. Casper, "The Impact of Economic Variables on the Demand for Library Services," American Society for Information in Science. Information Choices and Policies. Proceedings of the Annual Meeting, 1979, Minneapolis (White Plains, N.Y.: Knowledge Industry Publications, 1979), pp. 41-50. Evidence indicating elasticity can be found in M. K. Buckland, Book Availability and the Library User (New York: Pergamon Press, 1975).

¹⁰ For a detailed discussion see M. K. Buckland, Book Availability, summarized in M. K. Buckland, "An Operations Research Study of a Variable Loan and Duplication Policy at the University of Lancaster," Library Quarterly 42, no. I (January 1972): 97-106.

¹¹ This section is based on insights by Dr. A. Hindle, Department of Operational Research, University of Lancaster, England, during his participation in the work in the University of Lancaster Library Research Unit. For another account of this double feedback loop, see M. K. Buckland, "The Structure and Dynamics of Library Services," in Progress in Cybernetics and Systems Research, vol. XI, edited by R. Trappl, N. V. Findler and W. Horn. (Washington, D.C.: Hemisphere Publishing Corp., 1982): 147- 51.

¹² For an analysis of the relative importance of different reasons why readers cannot find books on the shelves, see P. Kantor, "Availability Analysis," Journal of the American Society for Information Science 27, no. 6 (October 1976): 311-9.

¹³ The double feedback loop concept presented here could be described as an example of multiple, more-or-less independent primary feedback loops modifying the basic system. This is quite different from the notion of a secondary feedback loop (or secondary controller) which modifies the first feedback loop rather than the basic system. If a thermostat is a primary feedback loop controlling a heater, then a secondary feedback loop would control the setting on the thermostat rather than the heater. For a library example of secondary feedback loop, see Buckland, "Structure and Dynamics of Library Service:" "The term variable loan and duplication policy describes policies whereby the length of the loan and the purchase of additional copies are made to vary as a function of pressure of demand. Hence, changes in the level of demand induce an adaptive response by the library. The library service becomes more responsive to demand with respect to the maintenance of standards of service—in this case, the probability of desired documents being available when sought. The librarian's role can move from a primary level of control (responding to needs as perceived) to a secondary level of control adjusting built-in feedback mechanisms to maintain whatever standards of service can be afforded (p. 149)."

¹⁴ See chapter 16 below. Also F. Narin and J. K. Moll, "Bibliometrics," Annual Review of Information Science and Technology 12 (1977): 35-58.

¹⁵ B. S. Nozik, "A Stochastic Model to Predict Demand for Library Services" (Ph.D. thesis, University of California, Berkeley, School of Librarianship, 1974. University Microfilms no. 70-18704); W. M. Shaw, "Loan Period Distribution in Academic Libraries," Information Processing and Management 12 (1976): 157-9.

¹⁶ V. Rosenberg, "Factors Affecting the Preferences of Industrial Personnel for Information Gathering Methods," Information Storage and Retrieval 3 (1967): 119-27; T. J. Allen and P. G. Gerstberger, Criteria for Selection of an Information Source (Cambridge, Mass.: MIT Sloane School of Management, 1967) (PB 176 899); 1. W. Harris, "The influence of Accessibility on Academic Library Use" (Ph.D. thesis, New Brunswick, N.J.: Rutgers University, 1966. University Microfilms order no. 67-5262); E. B. Swanson, "Information Channel Disposition and Use," Decision Sciences 18, no. I (Winter 1987): 131-45.

¹⁷ Buckland, Book Availability and the Library User.

¹⁸ For example, An Architectural Strategy for Change: Remodeling and Expanding for Contemporary Public Library Needs. Proceedings of the Library Architecture Preconference Institute, New York, 4-6 July 1974, edited by R. M. Holt (Chicago: American Library Association, 1976), p. 85; M. E. Veblen, Giant Strides Since Andrew Carnegie: Creative Architecture in the King County Library System (Seattle: Shovey Bookstore, 1975), p. 58. See also American Libraries 11, no. 10 (November 1980): 623; 13, no. 5 (May 1982); 322; and 13, no. 6 (June 1982); 356.

¹⁹ E. S. Palmer, "The Effect of Distance on Public Library Use: A Literature Survey," Library Research 3 (1981): 315-54.

²⁰ Cf. B. C. Brookes, "The Foundations of Information Science. Part III. Quantitative Aspects: Objective Maps and Subjective Landscapes," Journal of Information Science 2, no. 6 (1980): 269-75.

²¹ The relationship between recorded use (usually loan records) and unrecorded use has not been adequately studied. There is some indication that the two sorts of usage correlate, e.g., A. Hindle and M. K. Buckland, "In-library Book Use in Relation to Circulation," Collection Management 2, no. 4 (Winter 1978): 265-77.

²² S. C. Bradford, "Sources of Information on Specific Subjects," Engineering 137, no. 3550 (January 1934): 85-6. (Reprinted in: Reader in Operations Research for Libraries, edited by P. Brophy et al. (Englewood, Colo.: Information Handling Services, 1976), pp. 170-4; and in Collection Management 1, nos. 3-4 (Fall-Winter 1976-77): 95-103); S. C. Bradford, Documentation (London: Crosby Lockwood, 1948).

²³ For additional reading on scattering, see E. A. Wilkinson, "The Ambiguity of Bradford's Law," Journal of Documentation 28, no. 2 (June 1972): 122-30; Narin and Moll, "Bibliometrics."

²⁴ B. C. Brookes, "The Growth, Utility, and Obsolescence of Scientific Periodical Literature," Journal of Documentation 26, no. 4 (December 1970): 283-94.

²⁵ H. H. Fussler and J. L. Simon, Patterns in the Use of Books in Large Research Libraries (Chicago: Chicago University Press, 1969).

²⁶ For further reading on decay, see M. B. Line and A. Sandison, "'Obsolescence' and Changes in the Use of Literature with Time," Journal of Documentation 30, no. 3 (September 1974): 283- 350.

²⁷ See Palmer, "The Effect of Distance on Public Library Use: A Literature Survey," Library Research 3 (1981): 315-54.

²⁸ M. K. Buckland and A. Hindle, "Library Zipf," Journal of Documentation 25, no. I (March 1969): 52-7. This short paper is substantially extended by R. A. Fairthorne, "Empirical Hyperbolic Distributions (Bradford-Zipf-Mandebrot) for Bibliometric Description and Prediction," Journal of Documentation 25, no. 4 (December 1969): 319-43. M. K. Buckland, "Are Scattering and Obsolescence Related?" Journal of Documentation 28, no. 3 (September 1972): 242-6. Note that D. de S. Price had independently posited a relationship between decay and hardness/softness: D. de S. Price, "Citation Measures of Hard Science, Soft Science, Technology, and Non-Science," in Communication Amongst Scientists and Technologists, edited by C. E. Nelson and D. K. Pollock (Lexington, Mass.: D. C. Heath, 1970), pp. 3-22.

Go to Chapter 11