College and Research Libraries The Configuration of Reference 1 in an Electronic Environment J Jane P. Kleiner Administrators often look at the centralization of library collections and services as a means of improving finances, space availability, and user access. The effect on personnel, however, is seldom given equal attention. This article reviews both the positive and negative impact of centralization on staff and service. It describes the benefits of the central mode for professional development. The move from de- centralized to centralized library service at Louisiana State University is dis- cussed, as well as the effect of automation on that configuration. • he centralization versus decen- tralization argument regarding the configuration of libraries and collections has been sub- jected to decades of debate in the profes- sional literature. Opposing views on aca- demic library organization began appearing in the 1930s and 1940s. Since then, numerous papers have been pub- lished addressing the question. Entire conferences have argued the pros and cons of the alternative models. Maurice Tauber described the issue a quarter of a century ago as ''a significant topic for periodic re-examination. ' 11 Keyes Metcalf reasoned: ''As long as there are universities with large libraries, the ques- tion of centralization or decentralization will be a live topic for discussion, and, if I am not mistaken, the question will never be settled permanently one way or the other." 2 Early library organization followed a highly centralized design. Administration in large academic libraries was mostly cen- tral until World War II. Administrative problems were created by expanding col- lections and staff, so that other organiza- tional patterns began to emerge. College and university libraries began organizing along departmental lines in the 1940s with work divided amon9 circulation, refer- ence, and cataloging. In 1947, Rose Phelps described a library organization form in which reference was provided in "a subject departmentation" mode within the central organization. Her study compared reference service organi- zational patterns in three large public li- braries. In it, she discussed the advan- tages and disadvantages of each model, concluding that subject organization pro- vided superior reference assistance to pa- trons, though she did point out that such designs usually required separate general reference departments for ready reference and information service. 4 During this same period and in the fol- lowing decade, librarians increasingly in- corporated the concept of function in de- signing libraries and providing services. Raynard C. Swank promoted the educa- tional function concept and its relation- ship to library planning. He subscribed to the divisional plan as the most education- ally efficient design. That plan's tenets di- vided all functions of the library, except administration and technical processes, into subject areas-social sciences, hu- manities, and science and technology. 5 Jane P. Kleiner is Head, Reference Seroices, at Troy H. Middleton Library, Louisiana State University, Baton Rouge, Louisiana 70808. 302 J. R. Blanchard carried the subject concept a step further by proposing that librarians in "divisional reading rooms" be subject specialists rather than reference or circula- tion librarians. 6 STAFF SPECIALIZATION While this debate was in progress, li- brarians were arguing for staff specializa- tion instead of subject specialization. Lyle pointed to Peyton Hurt's argument that reference librarians should specialize in the history and bibliography of a field through ·systematic study of the literature rather than through experience gained in working with a subject collection. 7 Lyle agreed with Hurt but also recog- nized the importance of specialization in university library work, particularly in dealing with graduate students and fac- ulty. He did, however, caution the profes- sion about excessive emphasis on the value of specialization when providing li- brary service to undergraduates. Lyle pointed to the importance of general refer- ence specialization over subject specializa- tion in dealing with the undergraduate population. He argued that that category of user was served more effectively when "the reference librarian has the breadth of knowledge and human understanding to feel a real interest in the inquirer's needs. Narrow concentration in a field does not necessarily contribute to this idea. While it is proper and undoubtedly important that the reference librarian have special interests-a diversity of interest is all to the good-a concentration of study in one field may impair his understanding of un- dergraduate students. " 8 In the 1960s and 1970s, interest soared in this pattern of decentralization as numer- ous separate libraries for undergraduates were established. Interest in that organi- zational mode has diminished in recent years due to a variety of factors: improved library instruction methods, declining fi- nancial resources, and reduced enroll- ments.9 Against this backdrop of debate, new ar- guments have emerged focusing on the impact of technology on library organiza- tion. At a 1983 symposium on the topic, Anne Woodsworth argued: "When to- The Configuration of Reference 303 day's technology is utilized fully, the issue of centralized collections will fade into ob- livion. It will no longer matter."10 Atkin- son agreed with that concept in his predic- tion that the academic organizational pattern of the future is likely to be decen- tralized.11 Edward Holley, however, raised the question of users' demands for ''better ac- cess to more and more materials" in re- sponding to the viewpoint that technol- ogy could result in decentralization. He urged careful consideration of ''service needs, the administrative organization, and the fiscal impact of this centralization vs. decentralization argument. " 12 Martell proposed a plan that addressed the question of meeting library users' needs-a library organized into small client-centered work groups with librari- ans operating at all points where "the li- brary interacts with its user groups. " 13 In this organization, each work-group mem- ber would perform a variety of library functions such as advanced reference, col- lection development, bibliographic in- struction, original cataloging, plus other types of information service. Martell as- serted that this model would be more re- sponsive to users' needs. TECHNOLOGY EFFECTS In The Electronic Library: The Impact of Au- tomation on Academic Libraries, Cline and Sinnott suggested a type of matrix man- agement as the organizational mode, the physical design being drawn along subject divisional lines. 14 Others offer few con- crete plans for the physical organization. Instead, they challenge librarians to view their environments as information cen- ters, knowledge bases, or communities of scholars, rather than as storehouses of books. Patricia Battin urged universities to ex- plore the joint roles of computer centers and libraries as information handlers. She recommended integrating information technology into the existing information system to preserve the linkages to the present knowledge base. ''The very diver- sity of scholarly inquiry and information needs requires in the electronic age an un- precedented degree of centralized, coordi- 304 College & Research Libraries nated linkages and compatibilities to serve that diversity and permit the autonomy necessary for productive and creative scholarship," Battin asserted. 15 Richard McCoy voiced a similar philoso- phy when he called for an end to campus information compartmentalization. He urged libraries to adopt a new and broader information role when he said, "Today's research library is and must be a gateway to the broadest world of scholarly infor- mation, regardless of the information's format or source. It must be a channel to machine-readable databases and to sound and image data, as well as to books and periodicals. It must be the facilitator for ac- cess to the combined information of pri- vate, for-profit resources, as well as those of the public and educational sectors. It must be a channel linking scholars with the local, national, and international re- sources required to support their best work." 16 Though the centralization-decentraliza- tion debate has been waged for more than half a century, a resolution seems no nearer today. Indeed, the added dimen- sion of electronic information in libraries makes the question more complex. This paper provides no immediate solution to the organizational dilemma. Instead, it proposes centralizing one library service: reference. It explores the advantages and disadvantages of the service and staffing aspects of a centralized model and focuses on the quality of information assistance and personnel that can be provided in that setting in an electronic environment. A number of arguments related to the broader centralization-decentralization is- sue and the impact of technology on the organization are applied to the discussion of the reference operation. DEFINITION OF CONCEPTS The ALA Glossary of Library and Informa- tion Science defines organizational centraliza- tion as 11 an organizational arrangement within a library system which is character- ized by consolidated collections and ser- vice points in one central facility with few, if any, separate libraries outside that facil- ity." 17 The glossary describes decentral- ization as 11 An organizational arrange- July 1987 ment within a library system which is characterized by numerous geographi- cally dispersed collections and service points. " 111 ''Different interpretations of central- ization in libraries and in the design of library services according to func- tion have existed through the years.'' Webster's Third New International Dictio- nary defines centralization as ''drawing to or gathering about a center or central point and/or bringing under one control. " 19 Both the glossary and the dictionary defi- nitions of centralization are applicable to the centralized reference model proposed here, because this model draws together the functions of reference desk assistance, bibliographic and library instruction, on- line and electronic services, materials se- lection, and interlibrary loan into one ad- ministrative and geographic division. It proposes a service unit that is a microcosm of the university library designed to fulfill the purposes Guy Lyle attributed to refer- ence more than thirty years ago. These were (1) to provide answers to inquiries requiring specific information; (2) to teach students to use the library; (3) to provide bibliographical and other research assis- tance; and (4) to locate and make available less commonly used materials not in the li- brary.20 Different interpretations of centraliza- tion in libraries and in the design of library services according to function have ex- isted through the years, but it is notable that the principles of centralization and functional planning have remained rela- tively stable. Michael Krenitsky's report on university library design posed several premises as justification for a centrally ori- ented plan. These included emphasis on the use of books and materials, the great- est accessibility for the largest number of patrons, provisions for altering patterns of use necessitated by curricula changes, en- rollments, and/or collections, and the abil- ity of library staff to work with maximum 1 efficiency. His premises also identified the chief function of the library as service to its clientele. 21 In planning a reference design, Lyle incorporated these premises in his statement that in a well-designed refer- ence service area, a student working on a paper should be able to find all his refer- ences in printed bibliogr~phies, reference books, and periodical indexes in one loca- tion.22 These premises are still valid for a user- oriented design, and they are reflected in the centralized model proposed here. This study, however, extends beyond the lim- its of printed collections and incorporates the dissemination of all available types of reference information in today' s elec- tronic environment as well. CENTRALIZATION AT LSU Meeting the needs of users has been a critical factor in the move from a decentral- ized library system to a more centralized service at Louisiana State University and Agricultural and Mechanical College. The University serves more than 35,000 fac- ulty, staff, and students and is the state's comprehensive center of higher educa- tion. Additionally, it is one of only nine in- stitutions in the country designated as both a land-grant and sea-grant institu- tion. Its instructional programs include approximately 250 subjects that lead to both undergraduate and graduate profes- sional degrees. There are eighty-nine mas- ters and fifty doctoral programs listed in the 1985-86 Graduate School Catalog. 23 Students, faculty, and staff represent- ing a broad range of subject areas make up the LSU user constituency. With the ex- ception of libraries provided for chemis- try, special collections, and the profes- . sional schools of law, medicine, and library science, all materials are housed in the Troy H. Middleton Library. The Middleton building today is a five- floor structure housing the bulk of a mate- rials collection approaching two million volumes. Reference Services, located on the main floor adjacent to the entrance, provides service to users in all subject ar- eas except business, which is housed with government documents on the basement level, including topics where much of the The Configuration of Reference 305 material is located in chemistry, special collections, or in one of the professional schools. Prior to 1958, the university libraries consisted of the main library plus ten branch libraries. In 1958, construction of a new three-story building was completed, and all branch subject collections, except chemistry, were moved into the Middle- ton Library. The law and medical libraries were, and still are, administered sepa- rately. For twenty years, reference service in Middleton was provided through three subject divisions: humanities, social sci- ences, and sciences. Few changes occurred until the mid- seventies, when automation made its first inroad through participation in auto- mated cataloging via the Ohio College Li- brary Center, now known as the Online Computer Library Center (OCLC). This was soon followed in the reference areas by the introduction of BLISS, the Biblio- graphic and Library Information Search Service. Further centralization was accom- plished in 1978 when the three subject di- visions, excepting government docu- ments and business materials, were integrated into one collection. At the same time, the divisional reference depart- ments were united to form Central Refer- ence, the largest unit of the Reference Ser- vices division. To keep pace with the growth of the library and the student body, two more floors have been added to the building, and the physical space allot- ted to the central service has more than doubled. Since 1984, under the leadership of di- rector Sharon A. Hogan, Middleton has been enhancing its information role by in- corporating current technology to extend its user services. This began with the in- troduction of the disk products Info Trac and CO-Disclosure to the reference areas in 1985-86. Meanwhile, negotiations were under way to automate the library using the fully integrated library system NOTIS, Northwestern Online Total Integrated System, developed at Northwestern Uni- versity. This contract was soon followed by another providing membership in the Research Libraries Group. and participa- 306 College & Research Libraries tion in the electronic Research Libraries Information Network (RUN). In Septem- ber 1986, the first NOTIS module became operational with the introduction of Middleton's Library On-Line Access cata- log, LOLA. Middleton, particularly the reference service units, was now actively bringing an electronic environment to li- brary users in a centralized setting. Within this framework, Middleton ref- erence librarians serve an average of 350 to 400 users daily. On peak days during heavy use periods, this count easily ex- ceeds 500. Of course, the number of pa- trons not needing assistance but using the collection and electronic services could double those figures. An accurate total for this measure is not yet available. The area, however, is heavily populated with a vari- ety of users during most service hours. These patrons have ranged from elemen- tary school children to the director of a So- viet library. Their age, race, education, na- tionality, technical expertise, and ability or inability to speak English are as varied as the problems they present. ADVANTAGES VERSUS DISADVANTAGES Obviously, there are advantages and disadvantages to centralizing the refer- ence operation in a large university li- brary. Arguments usually focus on the economic advantages rather than the quality of service since financial factors are often more tangible and beneficial from an administrative viewpoint. Centralization of reference services does result in dollar savings, and this is one is- sue seldom disputed in the organizational debate. In Hurt's article proposing staff specialization as a substitute for library de- partmentalization, he spoke of the ''costly duplication of personnel, equipment, and books which comes with departmentaliza- tion.''24 Swank, the divisional system pro- ponent, also cited the fact that "it is com- monly charged that the subject-divisional type of organization is more costly than the traditional types/' though he person- ally questioned that belief. 25 Immediate financial benefits can be real- ized by reference centralization from staff reductions and ceasing duplication of ti- July 1987 ties for multiple locations. LSU achieved both benefits ten years ago by uniting the divisional reference departments into one service point. Four positions, two profes- sional and two support, were assigned to other areas at that time. Savings also re- sulted as duplicate orders were canceled. It was no longer n~cessary to provide a common core of dictionaries, almanacs, and other sources to three locations. Fur- ther staff time was realized by integrating the division catalog files into one, elimi- nating the need to maintain three separate catalogs. In addition to these pluses, arguments also point to the fact that consolidation results in space gains. New offices may be created, other units relocated, and/or added storage space for materials made available. These bonuses were also appar- ent in Louisiana and provided for the eventual expansion of the business ad- ministration/ government documents ser- vice, more office space, and the creation of a study room for the visually impaired. Additionally, a sizable compact storage area almost half the size of the former so- cial science area was created. The financial and space savings are the usual arguments proposed when central- ization is being considered. Frequently, the need to add footage and reduce ex- penditures are the reasons for consolida- tion. Existing conditions may force admin- istrators into immediate reorganization to resolve these concerns. When this occurs, the question of the effects of centralization on staff and service become secondary. A library's staf( however, is an institu- tion's life force, responsible for collecting, organizing, preserving, and accessing a li- brary's materials. In effect, the staff cre- ates a library and provides the service es- sential to making it usable. Without that service, the information contained in any collection is difficult if not impossible to access. The importance of staff to a library's suc- cess demands consideration. Unfortu- nately, the literature on this aspect of cen- tralization is slim. As pointed out previously, research has concentrated on the centralization-decentralization ques- tion as it relates to the library user, collec- . .... J ... tion development, or technology. Thomas Watts advocated centralizing library col- lections, noting such advantages as con- venience, the consolidation of collections in one location, reduced expenditure, and improved interdepartmental communica- tions. 26 Michael Hibbard saw merit in Watt's argument and favored the effi- ciency aspects of centralization. 27 ''Hugh Atkinson opposed centraliza- tion and emphasized patrons' prefer- ences for 'more focused units of li- brary service.' '' In "A Brief for the Other Side," Hugh Atkinson opposed centralization and em- phasized patrons' preferences for ''more focused units of library service. '' 28 He cited Snunith Shoham' s literature survey as evidence of that preference. 29 STAFF/SERVICE BENEFITS None of these arguments, or those pro- posed earlier, resulted from an in-depth examination of the advantages or disad- vantages of centralization to a library's staff and services. Because of that lack, a consideration of points relating to these factors is in order. In brief, centralizing the reference staff and services can yield the following seven benefits: 1. Availability of a large professional staff to provide a high quality of service to users. 2. Extended hours of reference service beyond those usually provided by most · university libraries or by branch libraries. 30 3. Because of the sizable number of pro- fessionals, the unit can be staffed at all times with a librarian. 4. The large staff provides a greater va- · riety of subject expertise. 5. Such a staff allows for a broader range of talents and abilities supporting more effective staff specialization. 6. The greater numbers allow for more varied professional interests and activi- ties. 7. The sizable staff provides increased The Configuration of Reference 307 opportunities for in-service training and staff development. All of these benefits result in better ser- vice to the public. But, what are the disad- vantages? Let's consider five: 1. Professionals must be familiar with reference resources in many subject areas. 2. Online searchers must be acquainted with databases in a multitude of subjects. 3. Bibliographic instruction must be available in all subject areas. 4. Super subject specialists can seldom be developed because of demands from the numerous subject areas. 5. Sheer number of demands for assis- tance, complicated by the variety of pa- trons encountered in a centralized opera- tion, can be stressful. Because these five disadvantages can overshadow centralization benefits, they will be addressed first and accompanied by suggestions as to how they can posi- tively affect a central operation. EXPERTISE QUESTIONED The point questioned most frequently is the staff's ability to work well with refer- ence resources in numerous subjects. Centralized reference opponents empha- size the impossibility of knowing all the sources and of being capable of doing re- search in them. Attacks focus on the need for reference librarians to be subject spe- cialists first and information specialists last, if at all. They ignore librarians' study on the organization and structure of infor- mation and the research process. Profes- sional expertise in developing search strategies and relating them to appropri- ate sources is seldom considered. Further- more, the current impact of technology on staff and their ability to use it effectively to retrieve information is inadequately ex- plored. Patrick Wilson touches on the role of li- brarians in accessing information in his book Public Knowledge, Private Ignorance: Toward a Library and Infonnation Service Pol- icy. Wilson says, "The librarian is not a specialist in information in general, but in information about records. The librarian's job is a job of management of information- bearing objects, and the continually im- proved pe~formance of that necessary job 308 College & Research Libraries is a natural and reasonable goal for the fu- ture. " 31 A 1985 article discussing the effect of an online catalog on a reference staff at the University of Illinois added another di- mension to Wilson's argument by assert- ing, "The era of the 'holistic librarian' is here. ''32 The article points to the range of informational expertise needed to provide effective service in today' s electronic envi- ronment. The range extends from such specifics as the contents of database rec- ords and access points to the broad gener- alities of library collections and the auto- mated systems used to access them. Evelyn Daniel also explored the aca- demic librarian's new role as an informa- tion resources manager. She emphasized that ''librarians will have to be competent to work in new ways" 33 and pointed to the possibility of a consultative method akin to the online searcher/clientele search in- terview. All of these concepts have merit. Librarians must understand the organiza- tion of information to access it in a system- atized manner. This is basic to librarian- ship and should be part of one's professional education and experience. It is as essential to reference service as the knowledge of the reference process itself. Indeed, it is part of the process. Essential to realizing this philosophy, however, is having a collection extensive enough to permit full exercise of our pro- fessional abilities. Granted, reference li- brarians in a centralized service may never develop the in-depth knowledge of each tool that colleagues who work in a limited subject area do, but they will never be sub- jected to the limitations in resources en- countered in those restricted informa- tional environments either. We live in an interdisciplinary era, and academic librari- ans need the broad resources and support to grow in that direction. Therefore, a centralized reference ar- rangement, due to the scope of its knowl- edge base, provides a fertile environment in which to better develop expertise in in- formation handling. Such a setting, with its large staff reflecting a variety of aca- demic expertise and talents, offers more opportunities for enriched professional development while maintaining a reliable July 1987 quality of service for users. It is the envi- ronment needed to develop the holistic li- brarians and information resource man- agers needed by users at this point in our technological development. THE ONLINE SERVICE This same argument extends to the de- mands centralization places on the online searcher. Again, understanding the orga- nization and structure of, in this case, the databases is foremost to quality searching. But knowledge of the databases and struc- ture alone is not sufficient. Searchers must be able to manipulate that knowledge. Such skill requires certain personal quali- ties and attributes. Ann Van Camp in- cluded as attributes a logical mind, the ability to exploit success, a willingness to go beyond formal training, and a retentive memory. 34 Donna Dolan emphasized con- cept analysis and flexibility of thinking. 35 Most agree that searchers usually have good problem-solving abilities and are creative thinkers. Naturally, some online intermediaries do not have all the traits, but their development is to be encouraged for the benefits they bring to online access as well as for their potential enrichment of the centralized service. Learning additional databases is de- manding, but the payoffs to the individ- ual, the library, and the clients are consid- erable. Brian Nielsen cites these in his article on searching and librarianship, "Librarians may be working harder, and they are devoting more of their time to pa- tron interactions of a higher order. They spend more time with the subject litera- tures. They spend less time doing routine reference work and supervision. In their interactions with patrons they have found new respect, and have greater control over the interactions than they have over the bulk of traditional reference interac- tions. " 36 A centralized service is an ideal setting for encouraging this growth. The avail- ability of staff members provides the work force needed to handle routine reference work more efficiently. Additionally, con- solidation of the online service with refer- ence reduces the number of personnel in- volved in supervisory tasks. This affords searchers more tiine to improve their on- line capabilities, thereby increasing their individual expertise and enhancing their image with library users. Therefore, the centralization of the on- line service in reference extends the bene- ficial professional environment by neces- sitating that more databases be learned, which further improves online searching abilities. At the same time, the study of the databases increases knowledge of printed sources. These factors result in improving the professional capabilities and self- image of the librarians and enhancing that image in the university community. The final product in all instances is improved service to users. BIBLIOGRAPHIC INSTRUCTION The next disadvantage frequently al- luded to is the need to offer bibliographic instruction in all subject areas. Instead of a disadvantage, however, this works to a central operations' advantage also. Caro- lyn Kirkendall and Carla Stoffle point out the benefits to faculty relations provided by BI programs. 37 In their essay on instruc- tion, they cite Nancy Gwinn's study and her statement: "Building faculty relations-getting out of the library and into campus affairs-is still the key to building support for the library's instruc- tional program and other services. " 38 A centralized arrangement requires var- ied bibliographic instruction programs. At LSU, these include an undergraduate li- brary credit course; numerous one-shot lectures in freshman English classes as well as for agriculture, education, or jour- nalism graduate students; and physics, engineering, and other departmental fac- ulty seminars. These extensive BI pro- grams, like the online service, stretch the staff's professional experience and pro- vide beneficial results to librarians and li- brary users as well. BI program benefits result from the playing out of the following scene: a re- quest comes in for a petroleum engineer- ing lecture. It is assigned to a staff member who prepares an Infofile-a selective, an- notated bibliography on the topic-and . also provides an online demonstration at the presentation. Researching the topic, The Configuration of Reference 309 writing the annotations, and preparing the Infofile, plus reviewing the relevant databases, familiarizes the librarian with print and online resources in petroleum engineering, thereby enriching profes- sional expertise in that subject area. "Building faculty relations-getting out of the library and into campus affairs-is still the key to building support for the library's instructional program and other services." The lecture presentation and online demonstration improve the librarian's im- age with the faculty and students, result- ing in improved relations with that de- partment. An added benefit at LSU is the availability of the growing Infofile series to all students on request. Providing BI in the large variety of sub- ject areas demanded by a central service results in frequent repetition of the above scenario, improving faculty and student relations on a campuswide basis. This benefits the instruction program and the participating librarians and enhances the library's image. SUBJECT SPECIALIZATION In listing other factors considered disad- vantages to centralization, the limitation on developing the super subject specialist was cited. Central service detractors raise this point as one of the greatest drawbacks to the model. It should be noted that it is advantageous to librarians in subject- specific areas, or to advocates of that ser- vice mode, to exaggerate the degree of ex- pertise needed to provide effective academic reference service. Also worthy of consideration is the fact that today' s in- formation base is more interdisciplinary and has fewer boundaries than ever be- fore. The subject specialist may be too hampered by the limits built into the sub- ject collection itself to serve users fully in our changing environment. Equally relevant to this argument is Wilson's definition of the librarian's spe- 310 College & Research Libraries cialty as that of managing information re- ~ sources rather than the information it- / self. 39 Extensive subject specialization may be more desirable in special libraries in business and industry. Universities, as the name implies, are responsible for a universal approach to education, particu- larly in today' s society. The reference divi- sion in a university library, therefore, should provide services in keeping with that role. This is not to imply that subject exper- tise is unnecessary. It is required in uni- versities, but an engineering degree is not essential in providing reference service, searching, and BI for engineering clien- tele. What is necessary is familiarity with the engineering vocabulary, understand- ing how the information is organized, and knowledge of the relevant print and on- line resources. This can be gained through experience and on-the-job training. As F. William Summers states in his article ''Education for Reference Service,'' being conversant with the discipline is expected. How a reference librarian becomes con- versant with that discipline-whether through academic studies or experience- is less relevant. 40 Summers' article includes a discussion of the future of reference education and suggests incorporating the following changes into library school curricula: 1. Less emphasis upon specific sources and more on a ready reference core of ti- tles. He points to the increasing use of computers and the need to know database content and characteristics as well as search strategy. 2. Emphasis on group and interper- sonal communication skills. He asserts that future librarians need to combine ele- ments of the communicator-par- excellence with the knowledge of sources and the organization of information and how to access it. 3. Emphasis on the need for subject knowledge. He suggests alternative ways of acquiring this, including joint master's programs, sixth-year or advanced mas- ter's, or on-the-job training. 41 Though Summers' recommendations were made without regard to organiza- tional structure, his educational prescrip- July 1987 tion is particularly applicable for the cen- tral setting in the electronic environment. Knowledge of a coordinated core of titles, of computers, and emphasis on communi- cation skills all precede the need for sub- ject specialization in Summers' list. Possi- bly, the need for superior subject expertise in academic libraries has indeed been ex- aggerated. A study of the subject specialist ques- tion is certainly overdue. Until one is con- ducted, however, decisions can only be reached on this issue by examining the available evidence. Today's evidence seems to indicate that the electronic envi- ronment increases the need for the ''holis- tic librarian," 42 or the information re- sources manager43 rather than the traditional subject specialist. An examination of use of LSU's refer- ence services division appears to substan- tiate this growing need for information specialists. During fiscal year 1985-86, the Middleton staff handled 132,827 questions-more than 81,000 at the refer- ence desks and another 51,000 at the adja- cent information desk. The large majority of these questions required the assistance of a professional capable of crossing inter- disciplinary boundaries. The demand for subject specialization was more evident in the 998 intermediated subject searches conducted online during the same period. Of that number, at least 30 percent, possi- bly more, did not require any subject ex- pertise due to the general nature of the topics. "At LSU, it appears that our elec- tronic environment may be shaping librarians who specialize in informa- tion management first, then subject content.'' The advent of the online catalog in fall 1986 has not altered that stance. If any- thing, it has added credence to it. It has further increased the need for librarians to know more about the database, the rec- ords' contents, and how the information is processed. Our experience has been similar to that at illinois in that regard. At LSU, it appears that our electronic envi- ronment may be shaping librarians who specialize in information management first and subject content second. ADDITIONAL ADVANTAGES The advantages of centralization at LSU have far outweighed the difficulties de- bated in the literature. By far, the greatest advantage has been the large professional staff allowing staffing by a librarian at all times. Additionally, this has enabled Middleton to provide more hours of refer- ence service than 84 percent of the li- braries surveyed by Paula D. Watson in her 1983 study. 44 It also permits the staff to deal effectively with the number and vari- ety of questions and users coming into the central service. Librarians seldom work more than twelve hours a week at the desk, which counteracts the stress and burnout possible in busy reference areas. Additionally, the staff size provides a broad variety of subject expertise. LSU reference librarians hold bachelor's de- grees in such varying subject areas as jour- nalism, mathematics, psychology, earth sciences, and English. Four have subject master's degrees as well. Experience spans a gamut of subjects and also includes functional expertise. Background includes work in the humani- ties, social sciences, sciences, and technol- ogy and in both academic and special li- braries. Additionally, a wealth of professional experience is available, in- cluding varying lengths of service in Bl, cataloging, circulation, collection devel- opment, federal and state publications, in- terlibrary loan, and, of course, a broad range of reference service. Academic backgrounds, professional experience, and special talents and abilities of this scope require the large and varied staff that is more readily achieved in the cen- tralized setting. An added advantage for centralized staff is the opportunity for in-service train- ing and staff development available in a personnel unit of this size and quality. Training is particularly important to to- day's academic librarians, who must be experts in the technology as well as teach- ers and researchers. The <;onfiguration of Reference 311 Librarians at LSU are heavily involved in automation, which has necessitated considerable training. A variety of online services have been offered for a number of years, and recently the staff has been in- volved with the installation of NOTIS and the introduction of LOLA, the online cata- log. Woodsworth's idea that technology would obliterate the arguments favoring centralizing45 has not been the LSU experi- ence to date. Instead, the increasing use of technology has required a more highly trained staff, a goal which can be achieved most efficiently through centralization. Training is expensive, and departmental libraries increase that cost because of the administrative complications involved. Scheduling replacements to allow for training in branches or satellite facilities can be a problem. Middleton's central op- eration reduces that difficulty. Frequent in-service training programs are instituted with little disruption to staff or service, in- suring that user services are up-to-date, comprehensive, and professionally pro- vided. "The LSU experience indicates that technology presently requires more centralization in order to maintain well-trained personnel to operate the limited number of expensive techni- cal devices available for use." Middleton reference staff are currently involved in two regularly scheduled in- service training programs. The more ex- tensive of these does double duty in train- ing the unit's graduate assistants, who number ten, or more including the part- time assistants covering nights and week- ends. The graduate assistant program consists of weekly thirty- to forty-five-minute lec- tures on topics such as the reference inter- view, service policies, patron communica- tion, online systems and sources, and resources in the humanities, social sci- ences, and sciences. These lectures are provided by the staff's senior members 312 College & Research Libraries and are often enriched by special lectures prepared by graduate assistants holding master's degrees in various subject areas. Members of the reference staff also attend these presentations when they involve subject areas in which improvement is needed. This exchange of reference skills, experience, talents, and academic back- grounds is vital to the staff's continued professional growth and to the provision of quality service. Additionally, a series has been devel- oped specifically for the permanent staff. The training-in-progress sessions, or TIPS, are usually ten- to fifteen-minute presentations offered at biweekly staff meetings. They cover new reference sources, reference titles needing atten- tion, seldom-used materials needing pro- motion, and difficult-to-use items. In ad- dition to these, more formalized training and workshops are held throughout the year on the online systems and databases, NOTIS, RLIN, and microcomputer use and software. The increasing use of technology in aca- demic libraries has not resulted in a dra- matic move toward decentralization to date. At least there is no indication of that movement in the literature. The LSU experience indicates that tech- July 1987 nology presently requires more centraliza- tion in order to maintain well-trained per- sonnel to operate the limited number of expensive technical devices available for use. Admittedly, future developments, the increased availability of technically trained staff, and the reduced cost of elec- tronic equipment could radically change that picture. For the present, however, ac- ademic libraries must deal with the prob- lems of equipment shortages and an often inadequately trained staff. These may be dealt with most effectively in a central set- ting. Centralization of the collections, of pro- fessional expertise, and of the reference units has enabled the LSU libraries to pro- vide a high quality of service in this com- plex environment. It meets researchers' modern information needs by incorporat- ing something of the one-stop-shopping concept outlined by Battin in her descrip- tion of the Scholar's Information Center. 46 At the same time and in more physical terms, it has fulfilled Lyle's traditional cri- teria for a well-designed reference area by providing students a single place where they can research a paper and find all the needed references in bibliographies, refer- ence books, and periodical indexes. 47 REFERENCES 1. Maurice F. Tauber, ed., "Centralization and Decentralization in Academic Libraries: A Sympo- sium," College Research Libraries 22:327 (Sept. 1961). 2. Keyes D. Metcalf, Studies in Library Administrative Problems: Eight Reports from a Seminar in Library Administration (New Brunswick, N.J.: Rutgers Univ. Graduate School of Library Service, 1960), p .133. 3. Barbara B. Moran, Academic Libraries: The Changing Knowledge Centers of Colleges and Universities (ASHE-ERIC Higher Education Research Report, no.8, 1984), p.32 4. Rose B. Phelps, "The Effect of Organizational Patterns on the Adequacy and Efficiency of Refer- ence Service in the Large American Public Library," Library Quarterly 17:281-95 (1947). 5. Raynard C. Swank, "Educational Function of a University Library," Library Trends 1:37-47 (July 1952). 6. J. R. Blandchard, "Departmental Libraries in Divisional Plan University Libraries," College Re- search Libraries 14:243-48 (July 1952). 7. Peyton Hurt, "Staff Specialization," A. L.A. Bulletin 29:417-21 (July 1935). 8. Guy R. Lyle, The Administration of the College Library, 3d ed. (New York: Wilson, 1961), p.129 . 9. Moran, p.33. 10. Anne Woodsworth, "Decentralization Is the Best Principle of Organization Design Where It Fits," in "Centralization or Decentralization of Library Collections: A Symposium," The Journal of Aca- demic Librarianship 9:196-202 (Sept. 1983). 11. Hugh C. Atkinson, "A Brief for the Other Side," in "Centralization or Decentralization of Library Collections: A Symposium," The Journal of Academic Librarianship 9:196-202 (Sept. 1983). 12. Edward G. Holley, "Reaction to' A Brief ... ',"in "Centralization or Decentralization of Library Collections: A Symposium," The Journal of Academic Librarianship 9:196-202 (Sept. 1983). The Configuration of Reference 313 13. Charles R. Martell, Jr., The Client-Centered Academic Library: An Organizational Model (Westport, Conn.: Greenwood, 1983), p.67. 14. Hugh F. Cline and Lorraine T. Sinnott, The Electronic Library: The Impact of Automation on Academic Libraries (Lexington, Mass.: Lexington Bks., 1983), p.174-75 . 15. Patricia Battin, "The Electronic Library: The Electronic Library- A Vision for the Future," in New Information Technologies and Libraries, ed. H. Liebaers, W. J. Haas, W. E. Biervliet (Boston: NACO Scientific Affairs Division, 1984), p.203. 16. Richard W. McCoy, "The Electronic Scholar: Essential Tasks for the Scholarly Community," Li- brary Journall10:39-42 (Oct. 1, 1985). 17. Heartsill Young, ed., The ALA Glossary of Library and Information Science (Chicago: American Li- brary Assn ., 1983), p .38. 18. Ibid., p.68. 19. Webster's Third New International Dictionary of the English Language Unabridged (Springfield, Mass.: Merriam, 1976), p.363. 20. Lyle, p.129. 21. Michael V. Krenitsky, "Approach to a University Library Design," (Bryan, Tex.: Caudill, Rowlett, Scott and Associates Report, no.RR-13, 1958), p.6-7. (ED 014 860). 22. Lyle, p.132. 23. The 1985-86 Graduate School Catalog (Baton Rouge: Louisiana State Univ. and Agricultural and Me- chanical College, 1985), p.8-9. 24. Hurt, p.417. 25. Swank, p.46. 26. Thomas D. Watts, "A Brief for Centralized Library Collections," in "Centralization or Decentral- ization of Library Collections: A Symposium," The Journal of Academic Librarianship 9:196-202 (Sept. 1983). 27. Michael Hibbard, "Centralized Library Collections? Well, Maybe: A Response," in "Centraliza- tion or Decentralization of Library Collections: A Symposium, ' ' The Journal of Academic Librarian- ship 9:196-202 (Sept . 1983). 28 . Atkinson, p.200. 29. Snunith Shoham, "A Cost-Preference Study of the Decentralization of Academic Library Ser- vices,'' Library Research 4:175-94 (1982). 30. Paula D. Watson, "Reference Service in Academic Research Libraries" (Unpublished survey con- ducted by ALA/RASD Reference Services in Large Research Libraries Discussion Group, 1983). 31. Patrick Wilson, Public Knowledge, Private Ignorance: Toward a Library and Information Service Policy (Westport, Conn.: Greenwood, 1977}, p.121. 32. Beth S. Woodard and Gary A. Golden, "The Effect of the Online Catalogue on Reference: Uses, Services, anq Personnel,'' Information Technology and Libraries 4:338-45 (Dec. 1985). 33. Evelyn H. Daniel, "Educating the Academic Librarian for a New Role as Information Resources Manager," The Journal of Academic Librarianship 11:360-64 Oan. 1986). 34. Ann Van Camp, "Effective Search Analysts," Online 3:18-20 (Apr. 1979). 35. Donna R. Dolan and Michael C. Krenin, "The Quality Control of Search Analysts," Online 3:8-17 (Apr. 1979). 36. Brian Nielsen, ''Online Bibliographic Searching and the Deprofessionalization of Librarianship, '' Online Review 4:222-23 (Sept. 1980) . 37. Carolyn A. Kirkendall and Carla J. Stoffle, ''Instruction,'' in The Service Imperative for Libraries: Es- says in Honor of Margaret E. Monroe, ed. Gail A. Schlachter (Littleton, Colo.: Libraries Unlimited, 1982}, p.42-93. 38. Nancy E. Gwinn, "Academic Libraries and Undergraduate Education: The CLR Experience," Col- lege Research Libraries 41:10 (Jan. 1980). 39. Wilson, p.121. 40. F. William Summers, "Education for Reference Service," in The Service Imperative for Libraries: Es- says in Honor of Margaret E. Monroe, ed. Gail A. Schlachter (Littleton, Colo .: Libraries Unlimited, 1982}, p.166-67. 41 . Ibid., p.167. 42 . Woodard, p.343. 43. Daniel, p.360. 44. Watson, "Reference Service in Academic Research Libraries." 45. Woodsworth, p.199 . 46. Battin, p.213. 47. Lyle, p.132.