College and Research Libraries Research Notes 277 The Curious Case of the Library Building Lawrence Lieberfeld A great many academic libraries have been built over the past twenty-five years, and a great deal has been written about their planning and construction. What is curious is the inconsistency between what is written and built, on the one hand, and the actual pattern of use of these buildings on the other. There is an abstract, almost mythic quality to the writings and to the widely utilized planning guidelines, as if there were no lessons to be learned by re- viewing what happens after the buildings are occupied. The result of ignoring experience in planning and designing academic libraries has been a substantial waste of resources. In turn, waste of plant resources can serve to diminish the operating funds available to libraries. Perhaps that prospect, in to- day's financial climate, can bring about more cost-effective planning and even a change in attitude toward capital budget- ing. In this paper, the initial sections deal with what the author has observed during many years at more than forty college and university libraries. There follow com- ments on established planning criteria and a brief review of the literature. The ar- ticle concludes with a few questions and some very simple recommendations. THE UTILIZATION OF ACADEMIC LIBRARY BUILDINGS This article describes the levels of utili- zation of library buildings, as observed, monitored, and measured during the course of consulting studies. For the most part, these consulting engagements dealt with facilities planning, either for an en- tire campus or for the library alone. In some cases the focus of the study was on library staffing, systems, or finances, but included considerations of facilities data. Of the forty-two studies, thirty-seven* incorporated systematic surveys of reader station utilization, and twenty-three in- cluded detailed analyses of book-stack uti- lization. Together, reader space and stack space constitute 80 to 90 percent of library space. These two elements are discussed in turn below. READER SPACE . The number of reader stations (carrels or lounge chairs or segments of tables, etc.) is usually allocated via a formula or a ratio between the number of students at the in- stitution and the presumed proper num- ber of these reader stations. The ratios ap- pear commonly in public university or regents' guidelines and in ACRL' s For- mula C. t Even institutions that do not offi- cially utilize written guidelines are often influenced by knowledge of libraries planned on this basis or by generally held beliefs as to -what ratios are appropriate. On the face of it, the very concept of a formula applicable to large numbers or classes of colleges or universities is ques- *Of these thirty-seven colleges and universities, thirteen are liberal arts colleges, seven are public or private universities, four are medical schools, and thirteen are other university libraries, e.g., business or science libraries. All but three of the liberal arts colleges would probably be classified as ''strong'' or "honors-oriented" in accord with Formula C of the ACRL Standards for College Libraries, 1975. +Formula C of the ACRL Standards presents space planning criteria. Our understanding is that this formula is to remain unchanged, although a committee is currently reviewing other aspects of the document. Lawrence Lieberfeld is a management consultant, Peat, Marwick, Mitchell & Co., New York, New York. 278 College & Research Libraries tionable. Variables that influence the stu- dent's decision during the course of the day or evening to study or browse in the library include the nature of the academic program, the nature of the collection and the library's circulation policies, the char- acter of student residences, the academic calendar, and the location of the library building. Why should this configuration of factors be the same for any two institu- tions? If there were no other way to determine the number of seats required in an aca- demic library, then a formula might be jus- tified. However, there is a simple method that is both logical and useful in the great majority of cases. The essence of the prob- lem is demand, that is, the number of reader stations that will satisfy the needs of students. This demand can be ascer- tained by a field survey of the actual utili- zation of tables, carrels, microreading sta- tions, and so forth, except when the demand is equal or nearly equal to the supply of such facilities. Under such cir- cumstances, the survey data might not be valid because students intending to use a reading area might be deterred by the ap- parent crowding. However, if the de- mand, as evidenced by periodic and sys- tematic observation, is substantially less than the supply of the available number of stations, then the measure of that demand under normal circumstances constitutes a valid basis for planning. Therefore, we ask those institutions for which we serve as consultants to conduct such a survey, using a simple one-page form. Each form covers one day's reader station utilization in one library building. The hours of the day and evening are ar- rayed at the left and across the page are columns representing the principal rooms or floor levels that provide reader stations. Further detail can be incorporated in the July 1983 form, e.g., the distinction between carrels and tables or detailed data on index tables or microreading stations. The demand can be expressed both as a percent of enroll- ment and as a percent of available reader stations. The surveys typically run for three weeks. Thus, if the library were open for ninety hours per week, there would be 270 counts of readers using the facilities. Results of the surveys are in ta- ble 1, based on the single hour when the most reader stations were occupied. Various objections can and have been made as to the validity and significance of these data. For example, the time of year is important. We usually ask the librarian what time of year generates the maximum load. Responses vary; some mention the week before semester examinations, some an earlier week, when term-paper activity is presumed to be at peak. Although our surveys obviously do not always coincide with the perceived peak periods, we try to adjust the results accordingly. In some in- stances, we have had the opportunity to come back for a second survey at a later point in the school year. The forms are filled out by library per- sonnel, who make the rounds and count the number of seated readers once per hour. In a few cases, security personnel were assigned to this task, in other cases, work-study students, and in many in- stances, regular employees. Another comment sometimes made by librarians is that current data do not take account of the impact that can be made by a new building-that the quality and amenities of a projected facility will attract students who do not utilize the existing building. We have found this to be the case only rarely. As the principal genera- tor of library use is the academic program, there is no significant reason why a change in the library's physical environ- TABLE 1 RESULTSOFTHESURVEYS Liberal arts colleges Universities Medical schools Other Peak Use as Percent of Available Stations 33.8 44.0 53.8 55.6 Peak Use as Percent of FfE Enrollment 10.6 5.1 ment will modify the pattern of use. It is worth recalling that we are assum- ing that libraries should be planned to ac- commodate peak demand for reader sta- tions. The median rate of utilization in the studies cited was typically 50-70 percent of the peak, even when weekend counts were excluded. How typical of academic libraries are the conditions revealed in these studies? Ob- viously, we cannot say with certainty, but our belief, based on an extensive, nation- wide higher-education consulting prac- tice, is that they are entirely typical. Li- brarians may not agree and are often defensive on this issue. If the librarian ex- pects use of reader stations to reflect ACRL' s Formula C or state planning guidelines, he or she may feel guilty about or unwilling to accept lower utilization rates. The problem is not in the libraries but in the formulas and guidelines. SPACE FOR THE COLLECTIONS The principal issue to be dealt with un- der this heading is the number of bound volumes (books and bound periodicals) that may appropriately be housed per square foot in the stack space of an aca- demic library. One might expect that gen- eral agreement on such a number might be possible. But for various reasons, the number has been elusive, and the many articles written on the subject have not been based on adequate data. Our conten- tion is that more books may be housed per square foot than is specified in the ACRL formula and the state guidelines. The discussion will be limited to bound volumes, which occupy approximately 75 percent of the collection space in univer- sity libraries and a somewhat greater pro- portion in college libraries. It is important to understand the vari- ables that, in concert, determine the num- ber of volumes that can be accommodated per square foot of floor space. 1. The size of the average book in the collection is obviously a significant factor. Most of the controversy over the question Research Notes 279 How big is a book? relates to the thickness of the volume, but the other dimensions are also significant. Our surveys provide information on about 9 million bound vol- umes distributed among thirteen institu- tions. The average thickness varies; for all 9 million volumes the average thickness is almost exactly 1.33 inches or 1 /9 of a linear foot. It would be reasonable to assume that the higher the proportion of bound periodicals in the collection, the thicker the average volume. Yet the medical school libraries in our surveys conformed closely to the average of 1.33 inches. Information on the thickness of volumes in academic libraries has not been promi- nent in the literature. Metcalf, in 1965, 1 re- fers only to architects Wheeler and Githens' measurements of the Brown University library volumes prior to World War II. He concludes that working capac- ity should be equated with six volumes per linear foot. We are convinced that seven volumes per foot is a proper average for planning collections space.* 2. Less commonly examined is the width of the base of the book-stack unit. We have found virtually no instances where we could not recommend bases 163 /4 inches in width, except for oversized books. The prevailing use of 203/4-inch- width bases, or even wider fixtures, is counterproductive . 3. The width of the aisle is a controver- sial subject. For functional purposes, i.e., circulation of patrons, employees, and book trucks, a 30-inch aisle can be recom- mended, as opposed to aisles of 36 to 38 inches, which are noted in certain refer- ence works, such as Architectural Graphic Standards. The situation has become mud- dled recently by the 1980 ANSI standard, 2 which calls for 36-inch passages to accom- modate wheelchairs. These standards have the force of law for public university construction in many jurisdictions and also are incorporated in Section 504 of the Rehabilitation Act. The ANSI standard appears to impose a substantial and un- necessary penalty on colleges and univer- *For a 3 '0" (nominal) shelf: 1.33" x 21 volumes= 27.9"; 27.9" /35.0" = 79.8% or, say, 80% of absolute capacity. 280 College & Research Libraries sities that are planning additions or new li- braries and serves no useful purpose. The better way to provide for the patron in a wheelchair is via a page. In many cases, the wheelchair user will not be able to reach the top or bottom shelves, even though he may be able to traverse the aisle. 4. The last variable is the relationship between working and absolute capacity of the shelf, usually expressed as a percent- age. We believe defining work capacity as 80 percent of absolute capacity is reason- able (Metcalf suggests 84 percent). The working-capacity concept implies that more space will be available when, say, 80 percent of the existing shelving is filled. Many librarians do not have this kind of implied confidence in their institutions, so that using a lower percentage to express working capacity represents a kind of hedge against the uncertain future. How- ever justifiable such a position may be strategically, it does not reflect book-stack management requirements. When the survey data and observations described above are combined, it can be demonstrated graphically and arithmeti- cally that 19.1 volumes per square foot can be accommodated, based on 8-inch (nomi- nal) shelves, 7 in height, 24-feet-long stack ranges, and 36-inch aisles.* If we then fac- tor in Metcalf's figures on the incidence of oversize books, the number of volumes is reduced to 18.8. Even if we also make an allowance for possible inefficiencies in ar- chitectural arrangements, the least pro- ductive outcome we can reasonably ex- pect, in new buildings or additions, would be 17 volumes per square foot. In renovating old buildings one might be limited to 15-16 volumes per square foot or even less, if significant architectural aberrations are present. We have found that inefficient book- stack layouts are very common, so that 12.5 volumes per square foot can be, but certainly need riot be, the result. For ex- ample, we have seen on several occasions columns 24 feet on centers with ranges July 1983 spaced 6 feet 0 inches on centers. Capacity could be increased by 25 percent by add- ing one more range per structural bay, re- sulting in 4 feet 9112 inches center to center spacing of the ranges, substantially more than adequate. GUIDELINES AND FORMULAS Reference has been made earlier to guidelines and formulas; these planning tools have had status and widespread use in library and academic circles for many years, at least thirty years to this writer's knowledge. Yet I have never seen, despite many inquiries, any data or validation from either the state boards of regents, the ACRL, or any other library organizations. 3 The regents' staffs usually cite the ACRL as the source of their planning criteria. On the reader station issue ACRL For- mula C calls for ratios between 1:4 and 1:3, representing reader stations: FTE enroll- ment. The smaller ratio is applied to "col- leges wherein less than 50 percent of the FTE enrollment resides on campus," and the greater ratio is to be used for the "strong, liberal arts, honors-oriented col- lege." Our data suggest that (a) the 1:4 ra- tio is excessive and wasteful and (b) the distinction between the two kinds of insti- tutions does not have an analog in the de- gree to which students use reader sta- tions. Perhaps the "strong" liberal arts college has better residential accommoda- tions; we are convinced that the primary locus of study is the student's residence, in virtually all institutions. The public university or board of re- gents' guidelines generally specify the 1:4 ratio, but some calculate on the basis of 1:3 for graduate students, without consider- ation of whether graduate students are be- ing accommodated in laboratories or else- where in their departments. Some add dedicated, enclosed carrels for a percent- age of the faculty, irrespective of the fac- ulty office situation on campus. We have usually found that faculty carrels are the least utilized of all reader stations. Our surveys of reader stations encompass all *This discussion does not apply to the reference collection, where fewer volumes per square foot can be accommodated. users: students, faculty, and visitors. On the planning of book-stack space, in- stead of 17 volumes per square foot in new buildings, 12.5 volumes is the most com- monly proposed figure. Metcalf demon- strated the feasibility of 15 volumes per square foot, and there is even more reason to avoid waste now than in 1965. The ACRL stipulates 10 volumes per square foot for the first 150,000 volumes and then · a graduated scale until, for holdings above 600,000 volumes, 14.3 per square foot are allowed. Thus, for a library of 500,000 vol- umes the recommended average volumes per square foot would be 11.2. The rationale for the graduated scale is unclear. Why should the area per bound volume be different for a library with 150,000 volumes on one floor than for ali- brary with 900,000 volumes on six floors? For the remaining categories of space, Formula C and the state guidelines are at least as spendthrift. The ACRL and many of the states stipulate that ''other'' space, that is, space for the library functions other than collection space and reader space, should amount to 20 percent of the entire net area of the building. We doubt that such a condition exists and are confi- dent that there is no justification for it, ex- cept perhaps for some library with a very distinctive activity pattern, such as a rare book facility. In fact, estimating other space (the most important single category in the ''other'' configuration is space for li- brary employees) as a percent of collection space is irrational. For example, a college with a collection of 500,000 volumes would generate more "other space" than a college with 250,000 volumes, according to Formula C. Yet the college with the smaller collection may be accessioning at a higher rate and, therefore, maintaining a larger staff. The larger collection may sim- ply be a function of the age and history of the institution. AN OVERVIEW OF THE LITERATURE The subject of physical planning criteria has not been dealt with extensively by any of the authors of the several facilities plan- ning manuals for libraries except Metcalf. Following Metcalf's 1965 work, Ralph Research Notes 281 Ellsworth published in 1973 a Planning Manual for Academic Library Buildings. One chapter deals with space standards; how- ever, Ellsworth does not propose stan- dards based on his own direct experience as a librarian and consultant. Instead, he cites the planning criteria utilized by a number of sources (including this author). He expresses a number of cautions con- cerning standards and formulas. Ellsworth Mason, in Mason on Library Buildings (1980), devotes only a few lines to space planning. He calls for seating 20 percent of the students at one time in resi- dential institutions, 10 percent elsewhere. For shelving, his specification equates with about 14 volumes per square foot. Aaron Cohen, in Designing and Space Planning for Libraries (1979), devotes much more space to furnishings, mechanical and electrical equipment, and design fea- tures than to determining the size of the building. One of his examples specifies 17.5 volumes per square foot, although that example is based on 5 feet 0 inches center to center book-stack range spacing. Heritage on Microfilnt Rare and out-of-print ·titles and documents on 35mm silver halide microfilm. • French Books before 1601 • Scandinavian Culture • 18th Century English Literature • Victorian Fiction • Literature of Folklore • Hispanic Culture Send for catalog and title information today. ~~&~ 70 Coolidge Hill Road Watertown, MA 02172 (617) 926-5557 282 College & Research Libraries He does not deal with the determination of the number of reader stations. It is fair to say that the literature does not provide information on planning either reader or stack space that is derived from institutional experience. CONCLUSIONS What does it all mean? Are Formula C and the guidelines serious attempts to ex- press functional requirements? Are they simply bargaining chips to be used in the endless game of capital budgeting? Do they express the values of the institution or only of the library? Unlike the ACRL' s Formulas A and B, which are designed to, but rarely do impact the institution's oper- ating budget, Formula C and the associ- July 1983 ated guidelines have a strong influence on the funding process and its outcomes. What is the remedy? Simply eliminate Formula C and the guidelines. There is no reason why the ACRL cannot content it- self with the kind of statement of general principles concerning facilities that is used by the ARL. In place of the guidelines let each institution develop library building programs on the basis of its own needs, without any predetermination or artificial assumptions hovering over the process. It seems to us that the time has come when the "more is better" philosophy of physical planning in higher education should be abandoned. It is time to recog- nize that buildings are liabilities as much as they are assets. REFERENCES 1. K. D. Metcalf, Planning Academic and Research Library Buildings (New York : McGraw-Hill, 1965). 2. American National Standards Institute A117.1- 1980. 3. The ARL Standards for University Libraries (1979) do not quantify physical planning criteria, although reference is made to Metcalf, Planning. 1~ y . t A . ·t· L.b . : ~ cqutst ton 1 rartans ••••• t 1 I I : I I! I Now is the time to try the best source for: • any book in print • accurate invoicing • meaningful reports 1 I • rush order service I t • competitive discounts t • PLUS MANY OTHER SERVICES I ~~ CALLTOLL-FREETODAY 1-800-248-1146 l 1 In Canada & Michigan I i CALL COLLECT (517) 849-2117 OTHER SERVICES .. .include binding of paperbacks upon request, returning slips in books, standing orders, staff orders, and up to date information on open orders. the BOOK HOUSE J081ER8 SERVING LIBRARIES WITH ANY lOOK IN PRINT SINCE 1112 208 WEST CHICAGO STREET JONESVILLE . MICHIGAN 49250 SAN 169·3859 :