Executive  summaryChapter1Chapter2Chapter3


The Patterns of Slide Library Circulation:
A Study

Rosalie Lack

Although there is no guarantee that usage patterns in a digital environment will mimic those in an analog one, it is likely that a number of causal factors (coursework assignment patterns, student procrastination) will persist. Analog use patterns can provide helpful insights into the kinds of loads and patterns that we might experience in the digital realm. This chapter examines circulation patterns at four university slide libraries. The circulation statistics provide benchmarks with which to compare overall use of digital image libraries and indicate likely periods of heavy use. The data gathered here helps reveal peaks and valleys in usage to help managers of both analog and digital collections to better plan staffing and technical resources.

We know very little about circulation patterns for images used in a university setting. Individual analog slide librarians usually have enough cursory knowledge about their own previous circulation patterns that they can estimate times that they will be very busy and times where they will experience very little user traffic. But there have been few cross-institutional or empirical studies of these patterns.

In this chapter, after first examining some of the issues related to circulation in the analog and digital environment, the chapter provides a summary of the major findings from our circulation survey, including usage patterns, percentage of users by user group and percentage of usage by month. It then briefly discusses the methodology of this study and presents a profile of each of the four slide libraries. Charts and tables with the detailed analysis of the data are available in Appendix 5B—Data and Analysis.

The analysis of analog slide library circulation patterns can be very useful to those managing digital image collections. For a number of years to come, storage space, processing power, and bandwidth will continue to be of concern to those managing large collections. Knowledge about usage patterns can lead to a better balancing of resources and a better design of systems architecture. For example, knowing that most of the use of a certain collection of images comes from a particular building can lead to a decision to store that collection on a server in that building, eliminating network bottlenecks. The knowledge that certain collections are used heavily during the last month of the term can result in the decision to set up "mirrors" of those collections on multiple servers, or a temporary shift of those collections from slow storage devices onto faster ones-easing bottlenecks in transferring very large images from storage devices. It is important to note that some technical solutions, such as replicating images on multiple servers, may violate licensing agreements or copyright protections.

For planning and design purposes, university digital image collection managers may also want some insight into the total use they may expect. Circulation statistics from analog slide libraries offer the only real evidence of the current extent of use of images in a university setting. Of course neither the extent of use nor the use patterns can translate directly into the digital world. (Some predictions suggest that the extent of analog slide use is likely higher than corresponding digital image use will be, because analog collections currently have the kind of coverage and critical mass that will be unavailable in digital collections. Others suggest that digital images will enjoy greater use, because digital systems are more easily accessible than slide libraries and thus will attract new audiences.) Nonetheless, patterns of analog slide use can give digital collection managers insight into the extent of use and usage patterns that they may experience.

Rudimentary knowledge of circulation patterns has led some analog slide libraries to purchase multiple sets of the same images in order to accommodate multiple simultaneous users at peak periods. Because analog collections follow the "serial use" model (where one person cannot use a work until the previous person has finished using it and returned it), managers of digital collections (which by their nature support "parallel use" models) may think that understanding multiple simultaneous use is not valuable to them. However, there are two areas where this understanding may be very relevant to those managers. Some digital licensing agreements will only allow access by one user at a time, so images that experience heavy simultaneous use may require extensions to the licensing agreements. And large digital images that experience multiple, simultaneous requests to view them will cause queues and major delays. Managers who are aware of these problems may implement architectures that minimize them.

Summary of Findings

Previous studies and surveys have looked at other aspects of slide libraries 1 ; this cross-institutional study focuses on circulation statistics only. We examined four analog slide libraries across a five-month period (January-May 1997). For the slide library that is at a university on a quarter system, we gathered data through June 1997. The libraries chosen for this survey were characteristic of cultural heritage academic slide libraries. Three of the slide libraries have a collection size ranging from 200,000 to 290,000 slides while the fourth has 366,000 slides. Two of them serve both the Art History and Architecture departments, one primarily serves the Architecture department and on serves the Art History department.

Despite the differences in collection type and size, there was a similar pattern of peaks and valleys in the circulation numbers (to view charts, tables, and further comments regarding the data gathered, turn to Appendix 5B—Data and Analysis). At each of the slide libraries, there was a pattern of at least two peaks during the semester or quarter. April was the busiest month for the three slide libraries on the semester system. For the slide library that was on the quarter system we looked at two quarters; they experienced high usage during two months, February and May. As might be expected for all the universities, the first and last months of the term were the slowest periods.

Data by type of user was only available at three of the slide libraries. The data demonstrated that for libraries that serve primarily one department, that department was, not surprisingly, their largest user. At the libraries that serve both Art History and Architecture, the Art History department was their larger user.

In addition, our study revealed a small but significant group of analog slide users that come from outside the designated slide library community. That is, the percentage of usage by secondary and "other" users was over 15% at all three slide libraries. We can expect that these numbers will increase in a digital world where gaining access does not require physically visiting an analog slide library located in a particular academic department. There was no clear pattern to the usage by these secondary and other users across the slide libraries. These findings reinforced the fact that courses in departments with slide libraries often tie their curriculum closely to the slide collection and use it on an ongoing basis. In contrast, other departments only use slides for a particular course segment or project, and therefore their usage patterns are sporadic.

One observation we made regarding individual usage was the ratio of number of slides that were checked out to the number that are pulled out and not borrowed. Users normally do not consult a catalog, but instead browse directly through the collection in order to find the appropriate slides. 2 The abilities to browse the collection and to juxtapose slides on a slide table are necessary functionalities in the analog world in order to help the user to compare images and find the precise image they need. Currently, these capabilities are not widely available in the digital world.


This study does not purport to be a comprehensive circulation study—it only examines four slide libraries. However we feel that these libraries are characteristic of cultural heritage academic slide libraries as a whole, and that these circulation patterns can provide valuable insight. We also believe that presenting these statistics is informative because although most slide libraries tend to keep their own statistics, there is a dearth of studies that compare the statistics across institutions.

The sources of the circulation statistics were logs maintained by the various slide libraries. The statistics were gathered during the spring of 1997. This study attempts to capture the dynamics of a slide library across an academic unit of time (semester/quarter), thus allowing us to examine the loads placed on a facility by an academic calendar.

Gathering circulation statistics for slide libraries is difficult because there are so many variables. For example, unlike a book library, slide libraries tend to have more informal rules regarding checking out materials. This is due in part to the fact that many slide libraries are in individual departments rather than under the umbrella of the main library system. Many slide libraries have policies that do not require that faculty check out material or they allow faculty to use the library outside of regular hours without accounting for the slides that they take out. Another factor that adds to the difficulty of assessing the data is that "circulation" refers to not only the slides that are checked out, but also slides that are examined in the library but not borrowed. Unfortunately, the number of slides examined in the library is not always logged.

In this study, there were various local environmental issues that made the data non-uniform across institutions, such as the fact that three of the universities were on the semester system and one was on the quarter system. However, we intentionally chose to look at a variety of institutions because it gave us the opportunity to examine and compare their strengths, weaknesses, and differences. We took these differences into consideration when making comparisons between institutions.


In this section we present brief profiles of each of the slide libraries (see Appendix 5A—Profiles of the Slide Libraries for a more detailed description of each slide library). Table 1 presents the comparison of the four libraries, by size of collection, number of users served per week and the departments that are served by them. Since it was beyond the scope of this article to conduct a detailed analysis of each library, we felt that it was not important to identify the libraries. For this reason, we have assigned each library a number code. 3

All four libraries have a number of digital images. At Slide Libraries 1 and 4, digital images are available across a network. Slide Library 1 offers digital images organized by course, while Library 4 is conducting an ongoing project to digitize all new acquisitions. Slide Libraries 3 and 5 offer digital images, but they are only accessible through workstations in the libraries. Due to the small data set (only 4 libraries), it was difficult to speculate on any correlation between the existence of a sizable digital collection and circulation statistics.


This circulation study provides us with benchmarks for circulation patterns and usage of university slide libraries. This type of comparative information is helpful, not only to those wondering where they stand in relation to other libraries, but also for those considering digital projects. This study lends empirical support to circulation patterns that many university slide collection curators would intuit about their collections.

The data gathered for this study (see Appendix 5B—Data and Analysis) indicates that there are peaks and valleys of usage throughout the academic term. In addition, the study shows patterns of usage of slides by user group and by month and it identifies the busiest and the slowest time periods of the term. It also demonstrates that there is a significant number of users outside of the designated slide library community. The usage patterns for this category of other users are not predictable and ongoing like those of primary users of a slide collection. It is anticipated that with the increase in availability of digital images this existent "other" category will continue to grow. A related observation is that there has been a general trend toward the increase in use of images, as Kessler notes "the interdisciplinary nature of any number of academic fields has significantly widened the audiences for visual images" (Kessler, 1994).

In order to meet the needs of the current users and anticipated future users, a recommendation for future research would be to conduct an in-depth usage study by classes of users. For example, a beneficial study would be to track the usage patterns of faculty as compared to students. Our study was able to gather this data for only one of the slide libraries, so we have no cross-institutional data. But the circulation data from this library exposed a relationship worthy of further exploration. The faculty and student usage patterns exhibited an inverse relationship; that is, when faculty had high usage (at the beginning of the semester) the students had low usage and vice versa. A usage study could identify patterns such as these and could also identify the various ways in which the images are being used. Those responsible for system design could benefit from the findings of such a study.

Appendix 5A—Profiles of the Slide Libraries
Appendix 5B—Data and Analysis
Appendix 5CBreakdown of Week Numbers by Date




1. See Roy McKeown and Otter, 1989, this comprehensive survey covers slide libraries in the United Kingdom; through questionnaires, the authors gathered information regarding all aspects of slide collection administration and organization. Included among the information is the size and coverage of each collection, the type of organizational system used, the types of services offered, budget and funding information, staffing, and the level of automation at each institution. Another important study concentrated on identifying the uses of slide libraries and how the organizational structure of slide libraries meets or doesn't meet user needs (Bradfield, 1976). In the Fall 1998 issue of the VRA Bulletin will be the results of a circulation survey conducted by Eileen Fry (Fry, 1998). Back to text

2. In his chapter entitled, "How Art Students Use Libraries," Philip Pacey states succinctly how important browsing is for art students when he says, "the art student is a compulsive browser", they "use libraries as reservoirs of images, visual information, example and stimulus" (Pacey, 1985:53). Back to text

3. The number codes correspond to those used for the slide libraries discussed in Part III, Chapter 4, The Cost of Distributing Analog Images by University Slide Libraries. Back to text


The Cost of Digital Image Distribution:
The Social and Economic Implications of
the Production, Distribution, and Usage of Image Data

By Howard Besser & Robert Yamashita