College and Research Libraries By I. T . L I T T L E T O N The Distribution and Cost of Library Service WH A T P R O P O R T I O N S of personnel ex-penditures and staff are allocated to cataloging, acquisitions, circulation, reference, departmental libraries, special collections? It is an unusual librarian who knows the answer to this question. It is generally assumed that the staff is dis- tributed in such a way that library serv- ice is integrated adequately with the teaching, research, and extension pro- grams of the university. T h e achievement of this basic objective of university li- braries depends largely u p o n the ade- quacy and the quality of the staff as well as the way in which materials are organ- ized; the organization of materials deter- mines to a great extent the organization of staff and distribution of salaries and wages. A n analysis of this distribution will give not only a better understanding of library service, but also some indica- tion of the extent to which the basic ob- jectives of the library are being achieved. There is need for a valid comparative study of the distribution of staff and per- sonnel budgets among departments and functions of a large number of libraries. Such a study would be useful to individ- ual libraries in long-range personnel budget planning and in pointing to staff needs, strengths, and weaknesses. It may also suggest more efficient distribution and use of staff. There is another important reason why there is need for an understanding of the use of salaries and wages. T h e largest proportion of the annual budgets of uni- versity libraries is expended, not for books, but for library service and for the organization and acquisition of materials in the f o r m of salaries and wages. This Mr. Littleton is assistant to the librar- ian, University of North Carolina. is true for each of the 107 Class I libraries in the A C R L statistical compilation for 1954-55.1 One hundred and five libraries spent over half of their total income on staffing. This information has special sig- nificance at the present time because of the recently increased concern about the financial problems of libraries. T h e participants at the Monticello Conference of the Association of Research Libraries held in October, 1954, recog- nized the need for comparative studies. President Morey, of the University of Illi- nois, "expressed the opinion that valid comparisons are extremely difficult, yet that such comparisons are almost the only way to judge whether or not a cost is rea- sonable."2 Williams in the volume that summarizes the conference says, "Infor- mation is lacking at present o n many aspects of library costs as a whole. . . . Uni- versities and their libraries d o a number of different things at once, and no one it appears, knows how much of the to- tal expenditures goes into each func- tion."3 T h e main reason that such informa- tion is lacking is the difficulty of obtain- ing valid comparisons. T h e same opera- tions are performed to some degree in most university libraries, but they are not performed in the same departments or in the same way in all libraries. It is also difficult to ascertain costs of functions which are distributed over several depart- ments, such as reference service or cata- loging. Therefore, a strict comparison by departments would have little meaning. However, for some functions, such as ref- erence service, it is impossible to extract 1 Dale M. Bentz, "College and University Library Statistics," CRL, X V I I (1956), 58-59. 2 Edwin E. Williams, Problems and Prospects of the Research Library (New Brunswick, N. J.: The Scare- crow Press, 1955), p. 38. sIbid., p. 12. 474 COLLEGE AND RESEARCH LIBRARIES the f u n c t i o n f r o m other duties of a de- partment, and a comparison of depart- mental costs is, in many cases, the only solution. Description of the Study T h e lack of detailed comparative data became apparent when an analysis of the use of staff and personnel expenditures at the University of N o r t h Carolina Library was made. It was impossible to determine if the expenditures and staff for functions and departments were reasonable since there were n o data f r o m other libraries with which to compare them. As a result, it was decided to attempt an exploratory investigation of the distribution of staff and salaries and wages in a small sample of libraries. A letter was written to the librarians of 30 libraries with collections of more than 500,000 volumes asking if they w o u l d participate in such a study. Sixteen libraries furnished data. Each li- brary was asked to submit the f o l l o w i n g for the fiscal year 1954-55: (1) the number of staff members, broken d o w n by pro- fessional and nonprofessional, in each de- partment and (2) the amount of salaries and wages allocated to separate depart- ments. T h e librarians were instructed to submit data separately f o r operations not representative of the major f u n c t i o n of the department in which they were per- formed. For instance, if the mending and repair unit was administered under the circulation department it was reported separately and not as a part of the totals f o r the circulation department. T h e de- partments were grouped under six main divisions: general administration, p u b l i c services, technical services, special collec- tions, special and miscellaneous services, and special and departmental libraries. T h e definitions of each of these categories f o r the purposes of this study are as fol- lows: General Administration includes the chief librarian, associate and assistant librarians, administrative assistants, secretaries, typists attached to the director's or librarian's office. It includes only those who participate in gen- eral administration and does not include per- sons who do specialized jobs, such as photo- duplication or interlibrary loans. Public Services refers to general lending and reference services, including the refer- ence, circulation, and documents depart- ments, undergraduate, graduate, reserve, and divisional reading rooms. It includes inter- library loans, but not extension service. Technical Services embraces the depart- ments of acquisitions, order, serials, catalog- ing, and mending and repair. "Cataloging" includes all persons who do complete cata- loging of books and serials regardless of the department in which they work, but not manuscript processing or public documents check-listing. Serials catalogers are included under "Cataloging" even if they work in the serials department. Special Collections includes separate col- lections of special materials, such as manu- scripts, maps, newspapers, rare books, prints, and state, university or archive collections. Special and Miscellaneous Services includes the following departments: extension, photo- duplication, mailing and shipping, and ren- tal collections. Special and Departmental Libraries: Data for each special or departmental library staffed with full or part-time personnel who devote their time exclusively to the library are included. Departmental secretaries in charge of libraries as a part of their duties are not listed. T h e 16 libraries in the sample range in size f r o m 2,696,862 to 552,171 volumes and have a regional distribution as fol- lows: Midwest, 5; Far West, 4; East, 1; and South, 6. T h e libraries were careful to report data within the limits of the definitions. T h e r e f o r e , the data, especial- ly for the broad divisions, have high va- lidity and comparability. In T a b l e 1, the p r o p o r t i o n of total sal- aries and wages and of staff assigned to the six m a j o r divisions as well as catalog- ing and acquisitions f o r each of the 16 libraries may be seen. T h e libraries are NOV EMBER, 1956 475 TABLE 1 P E R C E N T A G E S O F S A L A R I E S A N D W A G E S A N D O F F U L L - T I M E S T A F F A L L O C A T E D T O E I G H T D I V I S I O N S I N 16 U N I V E R S I T Y L I B R A R I E S , 1954-55 LIBRARY ADMINISTRATION ACQUISITIONS CATALOGING TOT. TECHNICAL PUBLIC SPECIAL SPECIAL SPECIAL SERVICES SERVICES LIBRARIES COLLECTIONS SERVICES Salaries Staff Salaries Staff Salaries Staff Salaries Staff Salaries Staff Salaries Staff Salaries Staff Salaries Staff & Wages if Wages & Wages & Wages £r Wages & Wages & Wages & Wages A 6.7 3.9 18.1 22.4 21.5 24.0 41.0 47.8 20.0 19.9 25.9 22.3 2.7 2.0 3.7 4.1 B1 4.6 4.3 7.7 10.3 16.3 18.6 26.7 32.8 15.6 16.6 46.1 41.5 4.2 2.4 2.8 2.4 C 10.1 7.0 10.8 12.9 24.5 27.5 36.6 42.7 27.8 24.6 16.7 15.2 6.4 8.2 2.3 2.3 D 9.9 5.9 19.8 21.3 18.8 18.9 39.5 41.4 20.0 21.3 24.0 26.1 4.2 4.1 2.4 1.2 E 5.5 4.25 18.5 18.2 21.6 25.4 40.8 44.25 25.6 22.4 19.6 20.6 4.7 4.25 3.8 4.25 F 5.6 4.4 16.5 20.4 16.7 17.6 35.2 40.9 28.0 23.4 17.3 18.1 11.6 10.2 2.3 2.9 G 9.4 6.5* 12.5 15.7 21.7 24.2 36.1 42.5 29.5 27.4 20.5 18.3 2.1 2.1 2.4 3.3 H 8.0 4.2 16.1 16.67 28.8 33.33 44.9 50.0 20.1 18.75 26.4 25.0 .6 2.08 I 11.2 7.0 19.7 25.6 8.9 11.6 28.6 37.2 33.0 25.6 24.7 27.9 2.5 2.3 J 4.5 3.0 16.6 17.7 19.5 14.5 37.6 44.2 27.8 22.5 26.6 27.7 3.5 2.6 K 9.5 6.7 11.4 12.1 17.9 21.5 30.3 34.9 22.6 22.1 13.7 12.8 13.8 13.4 9.1 10.0 L 9.6 5.8 16.7 20.4 21.5 21.3 38.2 41.7 28.8 24.3 15.6 16.5 6.2 9.7 1.5 1.9 M 5.8 4.1 13.4 16.4 19.1 22.1 34.9 40.6 27.5 23.5 26.9 23.4 2.9 2.8 1.9 5.6 N 10.4 7.8 19.6 23.3 21.8 24.5 45.7 54.1 24.7 20.2 12.6 10.9 5.8 5.4 .8 1.6 O 5.6 3.8 9.1 13.1 13.6 24.2 24.8 31.4 24.2 21.6 43.7 41.3 1.7 1.9 P 7.4 4.4 15.5 17.6 19.4 22.5 36.1 41.8 39.5 37.9 6.5 6.2 4.0 3.5 6.5 6.2 Mean 7.7 5.2 15.1 17.75 19.5 21.98 36.1 41.76 25.9 23.25 22.9 22.1 5.1 5.2 3.2 3.6 Median 7.05 4.35 16.3 17.17 19.25 22.3 36.25 41.75 26.2 22.45 22.25 21.4 4.1 3.8 2.3 2.75 High 11.2 7.8 19.8 25.6 28.8 33.33 45.7 54.1 39.5 37.9 46.1 41.5 13.8 13.4 9.1 10.0 Low 4.5 3.0 7.7 10.3 8.9 11.6 24.8 31.4 15.6 16.6 6.5 6.2 .6 2.0 .8 1.2 Number of Libraries 16 16 16 16 16 16 16 16 16 16 16 16 14 14 14 14 1 Data do not include the director and assistant director of libraries. arranged by size, the largest library first. "Acquisitions" includes all persons who participate in book and serials cataloging regardless of the department in which it is performed. The Use of the Data by an Individual Library T h e variations in the distribution of salaries and wages and of staff among li- braries is to be expected because of the differences in the sizes and interests of student bodies and faculties, in organiza- tional structure, in the sizes of campuses, in the traditions of institutions which tend to emphasize certain collections and departments, and many other variables. However, by studying these individual differences we can gain a better under- standing of the strengths and weaknesses of a particular library. For example, let us examine the profile of one of the libraries. Library K ranks thirteenth in the percentage of total sal- aries and wages and staff allocated to ac- quisitions, twelfth in the percentage al- located both to cataloging and public services, and near the bottom for the pro- portion spent on special and professional libraries. O n the other hand, in the per- centage allocated to special collections and special services, it ranks first. This in- formation, along with the comparative data on the actual number of employees, the actual expenditures for each depart- ment, and data on library size and total book budget can be used to determine where new positions need to be added. It can also be used to strengthen arguments for additional positions when presenting requests to university officials and the state budget bureau. Since the special col- lections and special services which draw such a high percentage of both staff and salaries and wages are heavily used, dis- tinguished, and well established, support cannot be withdrawn from them, but with these data, the librarian can explain why his budget request is as large as it is. T h e staffing problem in this library is not so much in the total numbers of person- nel as in a critical understaffing in general library services, public service and bib- liographical processing. A comparative analysis can also reveal weaknesses in internal organization. Transfers of operations from understaffed departments to departments more ade- quately staffed may result and weaknesses in departmental routines can be spotted. A n analysis of each of the other librar- ies will show different patterns of staff distribution. Each should benefit from a comparative analysis of its organization. Further Statistical Analysis By means of correlation technique, the data can be compared with many varia- bles to determine if there is any relation- ship between a particular factor and the organization and cost of staff. Some of the data that may be correlated with these are library use statistics, volumes added, total number of volumes, student enroll- ment, size of faculty, the number of aca- demic departments, and graduate and un- dergraduate degrees offered or conferred. T i m e has not permitted such a thorough- going analysis, but actual salaries and wages and the percentage allocated to each division were correlated with total number of volumes, student enrollment, number of special libraries and total sal- aries and wages by means of the Spear- man R h o correlation formula. T h e coeffi- cients obtained are not presented here since the sample is so small. Clear-cut pat- terns and trends were indicated by them, however. Some generalizations about the cost and distribution of library service can be made, but they are presented more as hypotheses than as conclusive findings. Many of the findings substantiate general assumptions held for many years. The Nature of University Library Service and Its Cost As a library grows in size its personnel budget and its staff grow also. This gen- erally confirmed assumption was proven NOV EMBER, 1956 477 statistically. In the sample there is a posi- tive correlation between library size and the total personnel budget and staff, but it is not a perfect correlation because some of the smaller libraries have per- sonnel budgets higher than libraries of greater size. These smaller libraries are growing faster and have a greater number of service units. Library M is thirteenth in size, but ranks fifth in total salaries and wages and second in the number of volumes added. This library also has three divisional reading rooms and an undergraduate library as well as a large group of departmental libraries. If the sample were larger, a study might be made of the staffs of libraries of compara- ble number of volumes. Not only does the total personnel ex- penditure of a library increase as it grows in size, but the costs of the various serv- ices increase also. However, all do not in- crease at the same rate. A smaller share goes into the staffing of public services and a larger share into an ever-increasing number of special and professional li- braries. Correlations between size and ac- tual expenditures for each of the divi- sions listed in T a b l e 1 were high and positive; but those between size and the percentage expended o n administration and public services were negative; and those between size and the proportion spent o n special libraries positive and moderately high. " T h e larger a campus grows and the larger the library grows, the greater is the inclination to split off portions from the central collection and transport them to locations more con- venient for the principal users thereof. Beyond a certain large size there is a de- sire on the part of users to flee from the large general services like loan desks and reference departments—operating in monumental Greco-Roman halls—to press for creation of smaller and cozier quarters and less impersonal service. This is g o o d but it certainly costs money."4 4 Williams, op. cit., p. 130. This keen observation by Coney at the Monticello Conference was substantiated by this study. As research strength in- creases, not only are more special libraries established, but they must be staffed with more highly qualified personnel. Many of the medium size libraries (between 500,000 and 1,000,000 volumes) have sev- eral departmental libraries which are staffed on a part-time basis by depart- mental secretaries, or not staffed at all, but the larger libraries and those that serve the larger universities have numer- ous departmental collections staffed with library personnel who usually have some knowledge of the subject. Size of library is not the only factor, perhaps not even the most important one that influences the splitting off of por- tions from the general collection. Some of the other factors that affect this relation- ship between general public service and departmental library service are: the re- search strength of the collection, the size of the student body, the number and types of professional schools and academ- ic departments offering graduate degrees and the arrangement of the library build- ing. Universities with the largest enroll- ments have usually developed more pro- fessional schools and graduate depart- ments. As this type of organization evolves, a larger percentage of the total personnel budgets go into staffing spe- cialized and professional school libraries, a smaller percentage into general public services. W h e n the total number of stu- dents was correlated with the percentage of the total salaries and wages allocated to public services, a very significant nega- tive coefficient (-.94) was obtained. T h e correlation between the total number of students and the percentage of personnel costs allocated to special and departmen- tal libraries was also very significant, but positive. There were high positive corre- lations between the total number of spe- cial libraries staffed and the total number of students (.96), whereas only a mod- erate correlation was obtained when the 478 COLLEGE AND RESEARCH LIBRARIES size of the library was correlated with the total number of special and departmental libraries. These data substantiate the theory that the development of a strong departmental library system depends more o n the size of the student enroll- ment, which to some extent reflects the n u m b e r of professional schools and aca- demic departments, than u p o n the size of the library. However, size in volumes is also a factor in these costs. A n o t h e r indication of the importance of student enrollment o n the develop- ment of special libraries is derived f r o m a comparison of the average percentages allocated to p u b l i c services and to special libraries in institutions of over 10,000 students and in those with enrollments under 10,000. T h e average percentage al- located to p u b l i c services in institutions with student enrollments of over 10,000 is 22.1, whereas 31.4 per cent is allocated to special and departmental libraries. In institutions with enrollments under 10,- 000 this ratio is reversed; p u b l i c service costs amount to 28.9 per cent and special and departmental library costs average 17.1 per cent. T h e most frequent special libraries in the sample are those that serve profes- sional schools. Fourteen of the institu- tions have engineering libraries, 13 have law libraries, and 11 have medical librar- ies. T w o of the medical libraries are ad- ministered separately and data f o r them are not reported. A large number of col- lections devoted to the sciences, especially in the fields of chemistry, geology, physics and mathematics, have developed in the institutions offering extensive graduate training in these fields. T w e l v e of the li- braries maintain separate chemistry col- lections, ten have geology libraries and nine have mathematics and physics col- lections. Other fields for which special li- braries have developed frequently are fine arts—especially art, architecture, and music—business administration, indus- trial relations, and education. Separate collections have developed less frequent- ly in the social sciences and humanities. T h e most costly special libraries are those devoted to medicine and law be- cause they are usually larger, must be staffed by personnel with specialized training, and demand long hours of open- ing. T h e average percentages of total salaries and wages allocated to these li- TABLE 2 MEANS, MEDIANS, AND RANGE OF PER- CENTAGES OF T O T A L SALARIES AND WAGES ALLOCATED T O LAW, MEDI- CAL AND ENGINEERING LIBRAR- IES IN 16 UNIVERSITIES, 1954-55 Law Medicine Engineering Mean 5.2 5.0 2.5 Median 4.8 4.3 2.55 High 11.2 11.1 6.5 Low 2.2 2.1 .9 Number of Libraries 13 9 14 braries as compared to those of engineer- ing libraries are given in T a b l e 2. A com- parative study of the staffing of libraries in the various fields w o u l d be a worth- while separate project. A l t h o u g h the actual amount spent o n administrative personnel increases as a library grows in size, the percentage of the salaries and wages allocated to gen- eral administration tends to be lower in larger libraries and in those with higher total personnel expenditures. Approximately one-fourth of the total salaries of these 16 libraries was expended f o r p u b l i c services. T a b l e 3 gives the av- erage percentages spent o n selected p u b l i c service departments. All 16 libraries maintain central circulation departments, even those which have divisional reading rooms. T h e p r o p o r t i o n allocated to circu- lation is less variable than that for any other department. Most of the libraries spend f r o m 8 to 10 per cent o n circulation department regardless of the n u m b e r of service units, the size of the library o r the size of the total personnel budget. A l l li- NOV EMBER, 1956 479 TABLE 3 MEANS, MEDIANS, AND RANGE OF PERCENTAGES OF T O T A L SALARIES AND WAGES AND STAFF ALLOCATED T O SELECTED PUBLIC SERVICE UNITS IN 16 UNIVERSITY LIBRARIES, 1954-55 CIRCULATION DOCUMENTS REFERENCE UNDERGRADUATE DIVISIONAL READING ROOMS READING ROOMS Salaries Staff Salaries Staff Salaries Staff Salaries Staff Salaries Staff