College and Research Libraries CHARLES J. POPOVICH The Characteristics of a Collection for Research in Business/ Management The citations from dissertations of Ph.D. candidates in the field of business/management were analyzed. The characteristics studied were publi- cation form, periodical title , subject, time span, language, and publisher. In addition, information was gathered that revealed the extent to which mate- rials were owned by the State University of New York at Buffalo (SUNYAB) Libraries. The study was based on 2,805 citations drawn from thirty-one dissertations-thirteen completed at S UNYAB and eighteen from other in- stitutions. ''E VERY LIBRARY exists chiefly to serve the needs of its own community of users. It follows, then, that any ... evaluatioh of a library [collection] ought to be based chiefly on how well it does, in fact, serve those needs." 1 One of the methods for gaining better insight to users' needs and a general perspective of the library collection is through checking relevant bibliographies that match the objective, purpose, and interest of the library and its clientele. In this study the approach employed was a citation analysis of bibliographies from dis- sertations. The two major aspecls of this in- vestigation were to: (1) define the charac- teristics of literature (interpreted as users' needs) cited by Ph. D. candidates in the field of business/management and (2) gain a getter perspective of the collection at the State University of New York at Buffalo (SUNYAB) Libraries in terms of ownership Charles ]. Popovich is business/economics li- brarian, State University of New York at Buffalo. The full report on which this article is based is available as an ERIC publication, "Business/ Management Research Characteristics and Col- lection Evaluation: A Citatio.n Analysis of Disser- tations" (ED 136 835). 110/ of materials cited in the dissertations . The resulting data are intended to serve as guidelines in defining a subject statement for the business/management collection. This information will ultimately constitute a portion of the overall collection develop- ment policy for the SUNYAB Libraries. BACKGROUND The SUNYAB Libraries system consists of eight unit libraries, five subunit libraries, and four storage facilities that collectively house more than 1. 6 million volumes of books and maintain subscriptions to 11,467 serial titles. 2 Although the majority of re- search materials for business/management are located at the Lockwood Memorial Li- brary (social science and humanities), it was decided this study would be conducted on the basis of the SUNYAB Libraries system. The rationale for this choice was the notable trend toward interdisciplinary research in the field of business/management, which undoubtedly would require the use of sub- ject materials from various 'library facilities. In addition, if the materials were owned by the SUNYAB Libraries, they should be ac- cessible regardless of location on campus. The specific questions concerning business/management research this study aims to answer are: (1) What is the form of publication most frequently used, i.e., peri- odical, monograph, serial, miscellaneous? (2) What are the most frequently cited peri- odical titles? (3) What is the subject disper- sion for monographs and serials? (4) What is the time span for the materials being used? (5) How extensive is the treatment of foreign language materials? (6) What are the most frequently cited types of publishers? (7) To what extent do the SUNYAB Librar- ies own the materials cited in the disserta- tions? A total of thirty-one dissertations were analyzed in an effort to find answers to these questions. PROCEDURE Thirteen of the dissertations that had been selected were completed at the School of Management at SUNYAB between Feb- ruary 1972 and February 1974. The remain- ing eighteen dissertations were those of in- coming faculty members who became affiliated with the School of Management at SUNYAB between September 1972 and September 1974 and finished their doctoral work at another institution. This latter group of dissertations, however, was com- pleted during various years, i.e., 1953 (1), 1961 (1), 1967 (1), 1969 (2), 1971 (1), 1972 (2), 1973 (4), and 1974 (6). A list of the schools where the faculty had completed their doctoral work and the number of dissertations that pertained were as follows: Berkeley (3), Carnegie-Mellon (1), Cornell (2), Harvard (1), Illinois (2), Kansas (1), MIT (1), Minnesota (1), NYU (1), Pennsylvania (1), Purdue (2), Rochester (1), and Southern California (1). The reasons for including the faculty dis- sertations were threefold: (1) the effects of a local situation on the research findings would be avoided; (2) the subject expertise of new faculty, denoted by the dissertations, may suggest current trends and emphasis in ( SUNYAB's business/management program; (3) it was assumed information from these dissertations would reflect the research interests of the faculty. The subjects. of the thirty-one disserta- tions were related to nine general areas of study, which included: accounting (6), eco- nomics (2), finance (5), health care manage- Characteristics of a Collection I 111 ment (1), industrial relations (4), marketing (4), organization and behavioral sciences (6), quantitative methods (2), and statistics (1). In recording information from the disser- tations, the following procedures were used: (1) If a source was cited two or more times in the same dissertation, it was counted only once. (2) If a citation was not a standard bibliographic reference, it was excluded from the study. Examples are a list of newspapers consulted-Oakland Tribune, Detroit Free Press, Baltimore Sun, etc., as well as references to cities where interviews were conducted-Dayton, Omaha, Seattle, etc. (3) Each reference was checked against the card catalog or the se- rials record to determine whether it was held by the SUNYAB Libraries. If there was any doubt about ownership of a particu- lar source, it was physically checked for lo- cation. (4) If a source was not owned by the SUNYAB Libraries but additional informa- tion was necessary, standard bibliographies and printed library catalogs were searched for the missing details. (5) The language of a cited publication was defined by the lan- guage of its title. (6) Classifying a citation by form was dependent , upon the manner in which the - author of the dissertation cited the material, and the investigator's judg- ment. As the citations from the dissertation bib- liographies were verified, the information was recorded on a twenty-column code sheet. This informatioh was ultimately keypunched into Hollerith cards. The Statis- tical Package for the Social Sciences (SPSS)3 program was used in tabulating and cross- tabulating the data on a Control Data Cor- poration (CDC) 6400 computer. The thirty-one dissertations produced 2,805 citations with an arithmetic mean of 90.5 and a range of 15 to 221 citations per dissertation. PREVIOUS STUDIES In analyzing the thirty-one dissertations (thirteen S UNYAB, eighteen outside SUNYAB), the method used was pioneered by Gross and Gross4 in 1927. Their tech- nique of "reference-counting" was an at- tempt to determine the most useful peri- odicals in the field of chemistry. Refine- ments of this study were performed by Fuss- 112 I College & Research Libraries • March 1978 ler when he investigated the use of serial and nonserial literature by researchers in the fields of chemistry and physics. 5 Another investigation, this time in the social sciences, was conducted by McAnally in his study of history literature. He used both books and journals published in 1903, 1939, and 1948 in an effort to determine whether there were distinct changes in the charac- teristics of materials used by researchers in the field of history. 6 In addition to these studies, there have been a number of investigations utilizing many similar procedures and techniques. The studies useful as a background for this particular evaluation were Stevens (library materials in doctoral research), Seagly (eco- nomics), Emerson (doctoral research in a university library), Sarle (business adminis- tration), Beckman (research collections), Intrama (public administration), Brace (li- brary and information science). 7 - 13 ASSUMPTIONS Although many of the previous studies argue the pros and cons of a citation analysis evaluation, it was not the intention of this paper to present a lengthy discussion on that aspect of the topic. The following as- sumptions, however, constitute the p~int of view taken in this particular evaluation. When analyzing the characteristics of lit- erature (users' needs) in a citation analysis study, it must be assumed a direct relation- ship exists between quantity (number of times cited) and the importance of mate- rials. Although this assumption is difficult to prove, investigators using this method have concluded that any defects that occur hap- pen so infrequently they are likely to have little effect on the accuracy of results. 14 In evaluating the SUNYAB Libraries' business/management collection, considera- tion of adequacy was based on the owner- ship of sources (perspective of the collec- tion) cited in the dissertations. This was in- terpreted as being supportive of business/ management research, and it was assumed that other doctoral candidates are likely to use the library resources in much the same manner. . Naturally, an assessment of where the collection is and where it should be going demands that the librarian be con- stantly alert to new courses, ·new programs, and new faculty int~rests. FORM An important consideration in allocating library funds is deciding what proportion of the budget to spend on periodicals, mono- · graphs, seri~ls, etc. Table 1 demonstrates for this study how publications were divided by form-periodicals, monographs, serials, and miscellaneous forms. An overall dis- tribution by form is shown first, followed by comparative data for Citations in disserta- tions prepared · at SUNYAB and outside SUNYAB. "A periodical is defined ... as a publica- tion. [other than a newspaper] issued at reg- ular intervals; at least twice a year. " 15 As table 1 indicates, periodicals constitute 49.1 percent of all the cited materials. A possible inference is that the field of business/. management is dependent on current re- search materials. Since periodicals provided that information, they were of primary im- portance. · Of the 1,377 citations to periodicals, 78 percent of the references (1,070 citations) were for materials in sixty-two different pe- riodical titles. A complete list of these jour- nals is included in the full report of this study. 16 TABLE 1 DISTRIBUTION OF CITATIONS BY FORM Overall Comparative Data Distribution SUNYAB Outside SUNYAB Form Number Percent Number Percent Number Percent Periodicals 1,377 49.1 604 50.8 773 47.8 Monographs 895 31.9 360 30.3 535 33.1 Serials 266 9.5 100 8.4 166 10.3 Miscellaneous Forms 267 9.5 124 10.5 143 8.8 Totals 2,805 100.0 1,188 100.0 1,617 100.0 Monographs were the second most fre- quently cited form of p~blication, accounting for 31.1 percent of the sources. A mono- graph was defined as "a single volume deal- ing systematically and in detail with a single subject. "17 This high ranking is not too sur- prising since monographs contain many basic studies and classic works. Those sources issued less frequently and regularly than periodicals were defined as serials. "The term includes ... annuals, numbered monograph series and the pro- ceedings, transactions and memoirs of societies. "18 With an overall distribution of 9.5 percent and a third place ranking, there is clear indication that this form of publica- tion is of lesser importance. The miscellaneous forms of publications, including doctoral dissertations, newspap- ers, reports, working papers, manuscripts, conference papers, master's theses, mimeographed documents, private com- munications, memoranda, as well as uniden- tified forms, represented a total of 9.5 per- cent. It is evident such forms are used only to a minimal degree and so were combined under the heading, "miscellaneous forms." The comparative data for dissertations from SUNYAB and outside SUNYAB show only minor differences between the two. Because of the relative consistency in the users' needs, as indicated by the form of publications, the budget allocation for these items at the SUNYAB Libraries was re- viewed . It was learned the business/ management acquisitions fund is weighted approximately 50 percent periodicals, 30 percent monographs, 18 percent serials, and 2 percent miscellaneous. 19 Although the SUNYAB Libraries' present spending pat- tern somewhat parallels the business/ management users' demands, perhaps a slight reapportionment may be necessary between serials and miscellaneous forms. SUBJECT DISPERSION The analysis of materials by subject could serve as a guideline in determining the ex- tent doctoral candidates in business/man- agement rely on various types of literature in their field as w"ell as outside their field. An awareness of these key areas allows for the selection of relevant materials and thereby an opportunity for providing better Characteristics of a Collection I 113 service ·to the library users. In this study the Library of Congress classification pro- vided the basis for defining subjects. The forms of publications analyzed were mono- graphs and serials. Generally, business/management mate- rials are included in the Library of Congress classifications HB through HJ. For the 1,161 citations pertaining to monographs and serials, table 2 shows a comparative dispersion of the ten leading Library of Congress subjects. They represent more than 75 percent of the subject sources in each category being compared, i.e., . SUNYAB, outside SUNYAB, and overall distribution. With the exception of HC (Economic His- tory and Conditions: National Production), the majority of business/management mate- rials (classifications HB-HJ) were from SUNYAB dissertations. Conversely, the majority of supporting literature (BF Psy- chology, HM-HX Sociology, etc.) was dominated by dissertations from outside SUNYAB. Information such as this is useful in helping to define the academic orienta- tion of SUNYAB's business/management program and in delineating the research interests of the School of Management fac- ulty. TIME SPAN "Time span may be defined as the extent to which the research worker in a given field reaches back into the literature of the past to find useful information. " 20 An awareness of the degree to which retrospec- tive research prevails in the field of business/management can aid in selecting library materials and in weeding the collec- tion . In this study the date of each disserta- tion was compared to the date of each cita- tion appearing in the dissertation. Table 3 indicates the distribution of time span by the fo;m of publications. More than 70 percent of all the cited ma- terials were ten years old or less, and nearly 85 percent of all the sources were cited within fifteen years. The rate of usefulness seemed to decline even more rapidly for the miscellaneous forms of publications during the first ten years. The results show there is a high obsoles- cence factor in business/management mate- 114 I College & Research Libraries • March 1978 TABLE 2 COMPARATIVE SUBJECT DISPERSION (IN PERCENTAGES) Outside Overall SUNYAB SUNYAB Distrihution perce nt of 460 percent of 701 percent of 1, 161 Rmk LC Classification Citations Citations Citations HF Business 17.4 13.0 14.7 5001-6351 2 HD Economics: Labor 4801-8942 19.4 6.4 ll.5 3 · HD Economics: Production 1-100 17.6 6.3 10.8 4 BF Psychology 5 .2 10.1 8.2 5 HM-HX Sociology 2.0 10.8 7.3 6 HG Finance 8.7 4.9 6.4 7 HB Economic Theory 7.2 5.3 6.0 8 Q Science 4.6 5.1 4.9 9 HC Economic History and Conditions: National Production 0.9 6.8 4.5 10 Political Science 1.1 6.6 4.4 --- --- Totals for leading LC Subjects 84.1 75.3 78 .7 Miscellaneous Literature 13 .5 18.0 16.3 Unknown 2.4 6.7 5 .0 Totals 100.0 100.0 100.0 TABLE 3 TIME SPAN OF CITATIONS BY FORM (CUMULATIVE PERCENTAGES) Form 0--5 6-10 Periodicals 42.7 72.8 Monographs 36.0 66.2 Serials 36.8 64.6 Miscellaneous Forms 53.6 76.1 Totals 41.1 70.3 rials , and recency of materials is a key con- sideration when building a collection or weeding it. FOREIGN LANGUAGE The extent to which foreign language ma- terials were ·used in business/management research revealed that it is practically nonexistent. Of the 2,805 citations that were analyzed, four , or 0.1 percent were to non-English titles. The distribution included two citations in French and one each in Spanish and German. O stensibly, the foreign language materials could he ignored, and it would he a tempta- tion to do so. However, the rise of multina- tional corporations ; the progres s of Japanese, German t and Russian technology ; advances of international practices in Years 11 - 15 16-20 21+ Unknown 87.0 93.0 100.0 80.7 89.5 99.9 100.0 82.3 89.1 98.9 100.0 88.5 90.4 90.5 100.0 84.7 91.3 99.0 100.0 business/management activities ; as well as other L'lctors may produce an added interest in international subjects. With these cir- cumstances, an increased proportion of writ- ings based on foreign literature may he forthcoming. At present, however , it ap- pears that doctoral candidates in business/ management make little use of research ma- terials written in languages other than Eng- lish. TYPES OF PUBLISHERS The decision to acquire certain library re- sources is sometimes based on the type of organization publishing the materials. To learn more about the publishers cited in business/management literature, the disser- tations were analyzed by categorizing the types of publishers into groups , i.e., com- Characteristics of a Collection I 115 mercial, association, government, and uni- versity. The commercial publishers were defined as profit-making organizations, such as Prentice-Hall, McGraw-Hill, Praeger, etc. The names of association publishers were established by consulting the Encyclo- pedia of Associations . 21 The definition of government publishers includes any monic- ipal, state, federal, or international agency of any government. 22 The university pub- lishers were references cited as " university press." In addition, references were grouped as "unpublished" and "unknown." Table 4 indicates the results. The totals indicate commercial publishers (33.8 percent) were the most frequently cited. They were closely followed by associ- ations (29 .6 percent) and then university (24.5 percent). References to government publishers were minimal (5.8 percent). In analyzing the types of publishers by form, periodicals issued by association pub- lishers were the most frequently cited (52.1 percent). A relatively strong representation was also indicated by university publishers (31.6 percent). The monograph form of publication was dominated by commercial publishers (74.1 percent), with university publishers a dis- tant second (18.2 percent). The serials form of publication was well distributed. The category most frequently cited, however, was government publishers (32. 7 percent). M