zhang.p65 Planning an Authority Control Project at a Medium-sized University Library 395 Planning an Authority Control Project at a Medium-sized University Library Sha Li Zhang Authority control is a vital part of providing students and faculty with adequate access to collections in university libraries. Although many large research libraries routinely maintain authority control, small and medium-sized university libraries find it challenging to meet rising user expectations and provide adequate access in an online environment through appropriate authority work. The smaller libraries have to pro­ vide persuasive data to demonstrate to their library administrations that tight staffing situations make it difficult to perform in-house maintenance on authority work and keep up with the dynamic and constant changes of new headings for bibliographic records. This article offers a planning process on an authority control project at a medium-sized university library. uthority control is a method that libraries must use to main­ tain consistency in their cata­ logs. As Robin K. Wendler stated: “Provision of consistent headings, together with a reference structure, allows searchers to isolate the particular author they are interested in and ensures that they find all available works by that author.”1 Henriette D. Avram wrote that authority control consists of the following elements: • distinguishing names: intellectual for­ mulation of the correct form of name fol­ lowing precedent and /or standard rules; • showing relationships: intellectual formulation of related names (variant forms, earlier or later names, parent bod­ ies, etc.); • documenting decisions: documenta­ tion of this information via creation of an authority record (thereby assisting sub­ sequent users of the same heading in de­ termining relationship and identifying headings on bibliographic records).2 In an online cataloging environment, authority control plays a more important role than ever before in enhancing user access to university library collections. Through adequate authority work, consis­ tent headings with a cross-reference struc­ ture enable users to locate needed library materials effectively in online catalogs. Background Avram pointed out that authority con­ trol is required as much as ever in online Sha Li Zhang is an Associate Professor and Head of the Technical Services Division in the Wichita State University Libraries; e-mail: shali.zhang@wichita.edu. The author wishes to thank the Office of Research Administration at Wichita State University, which granted a University Research/Creativity Projects Award to this project in December 1999. The author also wants to express sincere thanks to all cataloging staff at Wichita State University Libraries for participating in this project. 395 mailto:shali.zhang@wichita.edu 396 College & Research Libraries September 2001 catalogs.3 Users and librarians rely on the intellectual work of catalogers to distin­ guish items, works and versions, and headings. Although users are greatly as­ sisted by the collocation of materials, catalogers themselves, as well as acqui­ sitions and interlibrary loans staff, ben­ efit from authority control. The growing importance of authority control in academic libraries is evidenced by the considerable amount of attention given the topic in library literature in the past decades. Larry Auld’s 1982 review of eighty-year-old authority control lit­ erature from circa 1900 to 1984 included forty-six references and a list of seven additional bibliographic readings. The scope of his literature review was lim­ ited to publications in the United States and Canada.4 In the same year, in a semi­ nar paper on the topic presented to the Graduate School of Library and Informa­ tion Science at UCLA, Barbara Tillett in­ cluded 411 works. Except for thirteen, all of the works in Tillett’s citations were published after the 1960s.5 In Arlene G. Taylor ’s 1989 literature review concen­ trating on the philosophical aspects of authority control published in the 1980s, sixty-one works were cited, excluding those cited by Auld and Tillett.6 The author ’s recent search on the LibraryLit database, a subset of OCLC’s FirstSearch, revealed more than 450 entries that con­ tain the phrase “authority control” pub­ lished in library literature from 1984 to the present.7 Maintaining adequate authority work in university libraries is costly and labor- intensive. In large research libraries, the cost of maintaining authority control has been a part of routine expenditures on cataloging and database maintenance, although the levels of authority control vary from library to library. In 1979, au­ thority control cost ARL libraries ap­ proximately $5 million.8 Although maintaining authority con­ trol has been a routine activity in many large research libraries, being able to meet rising user expectations and provide ad­ equate access in an online environment through appropriate authority work has been a challenge for small and medium- sized university libraries. In her 1992 ar­ ticle, Joan M. Bechtel stated that “in an undergraduate academic library the most persuasive and responsible argument for providing authority control is that re­ trieval and use of information are signifi­ cantly enhanced. If the organization of materials is improved, that is, if the work of the catalog librarians is also enhanced by authority control, that is a bonus. Aca­ demic librarians, however, must not lose sight of the fact that their raison d’être is service to students and faculty, not their own comfort.9 Elaine Peterson and Bonnie Johnson estimated that it would cost $12,000 annually at the Montana State University Library to keep the authority file current. With this amount of funding, three hundred new titles would be pur­ chased. However, Peterson and Johnson were convinced that “if given the choice to buy 300 new titles which would be dif­ ficult to find or separated from other like materials in the catalog, or to bring all materials together into one place for the user by spending the money on author­ ity updating, the choice is clear. Once in an automated library, no patron is likely to leave the online catalog screen and use such tools as the LCSH red books (Library of Congress Subject Headings), even if the catalog is known to be deficient. Author­ ity maintenance is a must if patrons are to receive all information they need in one place.”10 The cost and staffing vary in maintain­ ing authority work, depending on the size of the library’s bibliographic records and the level of authority work it wishes to perform. It proves overwhelmingly that, with tight staffing situations, purely in-house maintenance work on author­ ity control is no longer able to keep up with the dynamic and constant changes of new headings for bibliographic records in small and medium-sized uni­ versity libraries. As Kerrie Talmac pointed out, “Although local systems may be able to make some global changes, staffing cutbacks and the attrac­ Planning an Authority Control Project at a Medium-sized University Library 397 tiveness of external agency services have led an increase in the number of librar­ ies looking beyond their own resources to obtain the required level of catalog­ ing integrity.”11 At the University of Day­ ton Library, another medium-sized uni­ versity library, Susan L. Tsui and Carol F. Hinders concluded that using a ven­ dor service for authority control “contrib­ utes to a library’s cataloging efficiency and helps improve the search capability and potential results for its users.”12 WSU Library Cataloging Environment Wichita State University (WSU) is one of three research institutions in the state of Kansas. The university also assumes the role of serving the educational needs in the metropolitan area of Greater Wichita. The university community includes about 15,000 students and 1,700 faculty and staff members. WSU Libraries cur­ rently holds more than 1.2 million vol­ umes of materials, with more than two million acquisitions budgeted for annu­ ally. Although the university offers ten doctoral programs and forty-five master ’s programs, most students who come to WSU enroll in a hundred and fifty undergraduate programs whose needs for research materials consist of both serial and monographic formats and both print and digital media. WSU Libraries automated its catalog­ ing system to NOTIS online catalog in 1985. Though the library’s users had ac­ cess to other libraries’ online catalogs through union catalogs, they depended heavily on access provided by the library’s online catalog, named LUIS (Li­ brary User Information Service), to the library collections. Each year, about 25,000 new titles are added to the library collections. As in the case of many small and medium-sized university libraries, the limited funding and tight cataloging staff prevented the library from fully implementing authority control at that time. Authority work remained in a very limited scope at the library. Authority records were created locally or exported from the OCLC Authority Files into the local catalog system only for music records, media records, and serial records. By November 1999, prior to the library’s system migration there were about 65,000 authority records in its NOTIS cataloging system cumulated since 1985. That was a very small frac­ tion, comparing more than one million bibliographic records in the cataloging system. Furthermore, these authority records had not been updated since the dates they were created or exported into the local catalog. Therefore, the changes in new, revised, and obsolete headings were not reflected in the local authority record file. The updated bibliographic records from the vendor were linked to more than 650,000 authority records in the new catalog system. In 1998, the library started importing bibliographic records in batch loads from a book vendor for both firm orders and approval books. This procedure repre­ sents 85 percent of new bibliographic records imported into the local catalog­ ing system on a monthly basis. The re­ maining bibliographic records came from OCLC’s WorldCat. The bibliographic records that are batch-loaded from the book vendor include full MARC records created by the Library of Congress (LC), CIP records, upgraded CIP records, and provisional records created by the vendor ’s cataloging staff. These biblio­ graphic records were entered into the library’s local system as they were, and no authority reviews were performed at that point. About 25,000 bibliographic records for new titles have come from this source annually. The bibliographic records imported from OCLC’s WorldCat were reviewed for valid head­ ings on a one-by-one basis, including music, media, and serial records. The headings for records created by the Gov­ ernment Printing Office and exported from OCLC’s WorldCat also are re­ viewed because the library has been cata­ 398 College & Research Libraries September 2001 loging all incoming government docu­ ment materials since 1994. The library has been receiving 62 percent of govern­ ment documents issued by federal agen­ cies since it became a federal deposit li­ brary in 1901. In the past decade, the li­ brary was able to maintain the stable ac­ quisitions of monographs that supported large undergraduate programs at WSU. With gift materials, almost fifty thousand new titles were added to the library’s online catalog in 1999 and 2000. Without adequate authority work, it was difficult to search and retrieve these valuable new collections. In November 1999, prior to the library’s system migration from NOTIS to Endeavor ’s Voyager online cataloging system, with one-time funding allocated by the library administration, the library contracted an authority control vendor to review and update all personal name, corporate name, series, conference name, uniform title, subject, and name/title headings for nearly one million biblio­ graphic records. During the review pro­ cess, obsolete and incorrect headings were replaced with valid and correct ones. The updated bibliographic records from the vendor were linked to more than 650,000 authority records in the new catalog system. This project enabled the library’s new cataloging system to be cleaner and more consistent than before. Students and faculty were pleased with the outcome brought forth by the project when they started using the new cata­ loging system in January 2000. Planning the Project When the library’s new cataloging sys­ tem was in full operation in January 2000, the cataloging staff began using an au­ thority validation tool of the cataloging system to review newly imported biblio­ graphic records against the existing au­ thority records in the database in the hope that all bibliographic records in the cata­ log would be kept up with heading changes. It was assumed that the head­ ings of newly added titles were reviewed and updated upon cataloging with the validation tool. If the headings for a bib­ liographic record were invalid according to this validation tool, the cataloging staff would search the OCLC Authority File and export appropriate authority records into the local online system to validate the headings of the bibliographic records. When an OCLC search failed to locate an appropriate authority record, local au­ thority records were created only for music, media, and serial records. No lo­ cal authority records were created for monographic titles that represented a large portion of newly acquired materi­ als. At the same time, the cataloging staff became more concerned with keeping up with the existing authority records in the library’s authority file. The LC issues 7,600 name authority records and 400 subject authority records every week.13 Though in many cases these headings may not affect the library’s bibliographic records because of the collection levels of the library materials, there is no auto­ matic method to identify a subset of these new authority records that do relate to headings in the library’s catalog. One cataloging staff member spent an entire week checking a weekly list of new head­ ings issued by the LC against the author­ ity records in the library’s local system. When an invalid heading was identified, the cataloging staff member would ex­ port an appropriate authority record from the OCLC Authority File and glo­ bally update all the invalid headings of bibliographic records in the local system. It was a very time-consuming process. Without ongoing authority work, library users would have a difficult time locat­ ing new titles. Therefore, the ongoing authority maintenance was brought to the discussion in the library’s Technical Services Division. This means that addi­ tional annual funding has to be allocated to have a vendor provide the service, as well as to pay for in-house staff time to maintain adequate authority work. During the discussion, the library ad­ ministration supported this ongoing commitment. The Technical Services Di­ � - �l � . 0: " : - l .. . r ' - < - : £ 0. " � 0: 0 0" Planning an Authority Control Project at a Medium-sized University Library 399 _ Q i Q i i Z §( i Q i i i i i 00 �( "9 8b � ; § 8'J ; t � " ,0 orI . ; t � �9" �( " . � ; § 8'J " , 0or I Z §( r00 �9" �( " . � '"" (!. 8( . � '"" (!. 8( " 0o0 I "9 8) �( .. ;'" ,; 6 " 0 o0I .. ;'" ,; 6 EE :( ""0 :E" ( " . I [1( [ 1 (V [:' " § V '1 ' 0 § §" [:' 0' § � I '( ' .(J '. ". .§.J ' '. '. § "E 8b �( . § § " r ;( \o0 �( ". §) (" . I ;"b (. � � �.. (" r. .(V � r. .(V . I ;"b )1 � � �.. (" r. .(V � r '." (V U ( ( g { �( .(; ( g { U0 �: r( §{ U0 8 . 8g( UU0 V( vision would have to ensure the proce­ dures that would merit the investment. The cataloging staff also looked into the following initiatives that would support the ongoing project: 1. The presence of technology: The li­ brary migrated to a new online catalog­ ing system in November 1999 that was client–server based. This new system of­ fers an authority validation tool, in ad­ dition to other modules such as acquisi­ tions, cataloging, circulation, and OPAC. Table 1 displays the tool. The authority validation tool informs the cataloging staff about invalid head­ ings, first-time-use headings that need to be verified, and obsolete headings that need to be updated when new titles are cataloged. This new feature has substan­ tially reduced the amount of time the cataloging staff has to spend manually on authority control functions. The new online cataloging system offers an excel­ lent opportunity to incorporate author­ ity control into the library’s routine flow of cataloging work. 2. The presence of the large number of authority records: When the library con­ tracted an authority control vendor to re­ view and update headings for all biblio­ graphic records in November 1999, the vendor also provided the library with 650,000 authority records. This procedure brought the library’s online cataloging to the most updated level. It was a bench­ mark for the library’s cataloging system. The cataloging staff in Technical Services Division wanted to grasp this excellent opportunity to maintain ongoing author­ ity control. 3. Redesigning staff structure: In Sep­ tember 1999, the Technical Services Divi­ sion redivided the existing cataloging workload, which comprised materials cataloging for commercially purchased materials, for incoming government documents, and for music and media ma­ terials. After redesigning the staff struc­ ture, three major cataloging tracks were created: Monographs Cataloging, Serials Cataloging, and Music and Media Cata­ loging. Three catalog librarians were as­ 400 College & Research Libraries September 2001 signed to each cataloging unit as resource persons. They provide procedural guide­ lines, make recommendations, and assist the cataloging staff in authority work. 4. Application for the NACO Music Project (NMP): In 2000, an application for the NMP submitted by the Music and Media Cataloger was accepted and ap­ proved. Through this funnel, the cata­ loger started training through an inde­ pendent and experienced NMP partici­ pant. The authority records for music materials created by this cataloger were then reviewed by the trainer. Subse­ quently, the revised versions of these records are contributed to the National Authority File. This catalog librarian also serves as a resource person on the au­ thority control project. Outcomes of the Project During the planning process, three types of records were examined: the number of monthly exported authority records, the sample of unauthorized headings, and the number of monthly batch-loaded biblio­ graphic records from a book vendor. Number of Monthly Exported Authority Records In March 2000, when the cataloging staff became familiar with the new online cataloging system and when the catalog­ ing work flow also became more stable after the system migration, the catalog­ ing staff was asked to keep as accurate a daily record as possible of the number of authority records that were searched from the OCLC Authority File and the number of authority of records that were actually found and exported into the library’s local cataloging system. Upon cataloging each new title, the cataloging staff reviewed each bibliographic record for such headings as name, title, subject, name/title, uniform titles, and series by using the authority validation tool of the cataloging system. If the heading was in­ valid, the cataloging staff would go to the OCLC Authority File to search and ex­ port corresponding authority records for the headings. If no authority record was available, a local authority record was created for a serial record, a music record, and a media record. No local authority record was created for monographic copy cataloging materials. The record- keeping part was very time-consuming. The cataloging staff felt much relieved when it was over at the end of the month. According to the daily record-keeping, the name heading consists of a large part of authority records that the cataloging staff exported from the OCLC Authority File during the month. In this category, 845 authority records were exported into the library’s local online cataloging sys­ tem. Nineteen local authority records for name headings were created. The next large category is the subject heading. The cataloging staff exported ninety-three subject authority records. For series records, seventy-six authority records were exported to the local cataloging sys­ tem. Four local series authority records were created. The category of name/title headings was a relatively small group. Only twenty-eight authority records were exported to the local database. Nine name/title authority records were cre­ ated. In summary, in March 2000, the cata­ loging staff exported a total of 1,049 au- TABLE 2 Summary of the Authority Record Activity in March 2000 Sources of Name Subject Name/ Title Authority Records Headings Headings Series Title Headings Headings Exported from OCLC 845 93 76 28 7 Locally created 19 0 4 9 7 Total 864 93 80 37 14 Planning an Authority Control Project at a Medium-sized University Library 401 TABLE 3 Descriptive Summary of Authority Records in 2000 (12 months) Min. Max. Sum Mean Std. Deviation Authority Records 320 1,049 7,721 643 216 Valid N = 12 thority records from the OCLC Author­ ity File to the local online cataloging sys­ tem. The total number of locally created authority records was thirty-nine. Table 2 summarizes activity. In table 3, the number of authority records (1,049) that the cataloging staff exported from the OCLC Authority File and locally created in March 2000 was much higher than the mean (643) in other months in 2000. The following factors contributed to this phenomenon: • Because twelve cataloging staff members in the Cataloging Department kept a daily record on what they did on authority records, their record-keeping was collected every day in March 2000. That function served as a reminder for the cataloging staff, who kept the author­ ity records activity in more accurate fash­ ion in March 2000 than in other months. Although the cataloging staff kept activ­ ity records at other times, those numbers were collected on a monthly basis. Of­ ten there was a tendency to forget the record-keeping. Therefore, the number of authority records reported by the cata­ loging staff during other months was less than the number they actually worked. • Three staff members from the Ac­ quisitions Department and the Serials Unit of the Cataloging Department also participated in some levels of mono­ graphic copy cataloging activity when monographic backlogs went up in the Technical Services Division. Because of the complex nature of the authority con­ trol procedure, these staff members were not asked to review authority headings upon cataloging during the year except in March. Therefore, a certain percent­ age of headings of new monographic records were not reviewed during the cataloging process in other months. The estimated staff time for review­ ing and exporting the appropriate au­ thority records is two minutes per record at the University of Dayton Library.14 At WSU Libraries, the cataloging staff re­ ported the similar pattern of performing this function at between 1.8 and 2.3 min­ utes per record, or thirty-six hours per month for 1,100 records. Sample of Unauthorized Headings The number of unauthorized headings generated through the local cataloging reports was examined to determine whether the in-house authority mainte­ nance was capable of keeping up with the newly issued authority headings by the LC. On a weekly basis, the system analyst who maintains the cataloging server at the WSU Computing Center runs several system reports for various headings. Table 4 displays samples of the number of incoming new bibliographic records whose headings were invalid; that is, there were no corresponding valid headings in the local authority file. Based on the reports, the cataloging staff had to export valid headings from the OCLC Authority File to validate all head­ ings in the reports. Because of the large number of headings generated on a weekly basis, only a few weekly samples were selected here to show the signifi­ cance. As is indicated in table 4, according to the authority reports generated by the online system, a large number of unau­ thorized headings was detected, includ­ ing unauthorized name headings, sub­ ject headings, name/title headings, and title headings. Because the system does not differentiate the genuine invalid headings from the valid headings that require some degree of editing/revising, http:Library.14 402 College & Research Libraries September 2001 TABLE 4 Sample of Weekly System-generated Unauthorized Headings in 2000 Week of Week of Week of Week of Week of Feb. Mar. Apr. May June 4-10 17-23 14-20 5-11 2-8 Unauthorized name headings 409 689 421 266 248 Unauthorized subject headings 421 854 375 367 347 Unauthorized name/title headings 725 751 529 456 306 Unauthorized title headings 21 48 22 17 23 it is very time-consuming for an in-house authority maintenance because the cata­ loging staff has to manually review each heading to decide when to export a valid authority record. Again, because the sys­ tem does not differentiate main subject headings from subdivision headings, there was no other way to obtain the ex­ act number of unauthorized main sub­ ject headings. If the number of unautho­ rized headings listed in table 4 was timed by four weeks, one could easily figure out the amount of staff time needed just to catch up with new headings detected by the local system each month. Number of Batch-loaded Bibliographic Records from a Book Vendor Finally, the number of newly imported bibliographic records from the book ven­ dor was examined. In 2000, the total number of batch-loaded bibliographic records from the book vendor to the lo­ cal cataloging system was 25,336, an av­ erage of 2,111 per month. The biblio­ graphic records from the vendor included both firm orders and approval books. As mentioned earlier, these bib­ liographic records were batch-loaded into the local cataloging system without being reviewed for their authorized headings. Table 5 lists the number of monthly batch-loaded bibliographic records. The original intention of batch-load­ ing bibliographic records from the book vendor was to speed up the acquisitions and cataloging process for newly pur­ chased monographic titles. That is, if an imported bibliographic record was com­ plete enough, the acquisitions staff would attach purchase orders and in­ voices to it. After that, the cataloging staff would simply add local holding and item records to the bibliographic record and would not go to OCLC’s WorldCat to search and export the same bibliographic record into the local system. Based on the library’s current cataloging procedures, the cataloging staff still needs to go to the OCLC Authority File to search and export the appropriate authority records upon cataloging the new titles. This pro­ cedure substantially slows down the en­ tire cataloging process. Additional Findings of the Project After examining the current authority control work flow, analyzing heading re­ ports generated by the cataloging sys­ tem, and reviewing the batch-loaded cataloging records, it became very clear that: 1. Additional staff time is needed to incorporate authority control into routine cataloging work flow, including time spent searching, reviewing, and export­ ing the appropriate authority records; checking the weekly list of new headings issued by the LC against the headings in the local system; and updating the exist­ ing authority records. Although the online cataloging system provides a function of global heading changes (see table 6), it still requires manual reviews and updates on each heading that is re­ lated to bibliographic records in the library’s catalog. The cataloging staff needs to manually click on the “Process” button, and the system analyst needs to run a global update program in the sys­ tem server to complete each update. � . � \ Z: e. .l 0. :. ::0 el� cc :u � .u0 l�' � .uc I:0 �� .� �l 0e � 00. . 5� 0l :5 N00 0 Planning an Authority Control Project at a Medium-sized University Library 403 �;i .O. �; . -c . �; . �:i O �: .. -: � �Oc t 0l t ZQ , "O l Q t;. § 0 "( V �. M N0 0 0o 0 tN t MN 0 N\ 0 0\ N \0 ' 00 ' 0t o 0t ' t0 t 0NN " " 0 ; � ,0 0�V \0' oM0 \M '00 t\0 'MM 'o\ 0 0M o0' \ 0 \t0 \00 t ' ' � .;" 0 MM N M "" N \ 0M N MN N o 0o o N0 t \ N o 0N t No 0 o N N \\ N 0 0\ No MM\ �O .: I N , ,, li .. �O ... i.: : I Mt o 2. The costs for in-house authority maintenance include the fees of export­ ing authority records from the OCLC Authority File, the staff time spent on maintaining new headings, and the staff time spent on updating the existing head­ ings. In their analysis on staffing cost, Susan L. Tsui and Carol F. Hinders con­ cluded that the cost for the total manual exporting authority records from OCLC, for staff time of reviewing and loading appropriate authority records, and for manual review of the existing authority records is more than having a vendor ser­ vice that “provides automatic machine correction of bibliographic records and supplies authority records with their re­ vised forms, leaving the library only the complicated database maintenance of its manual review and correction.”15 3. The level of commitment to the au­ thority control procedure in the library’s Technical Services Division includes: • Head of Technical Services Division: Planning and coordinating work flow and assignments for incorporating au­ thority control into routine cataloging work flow; • Cataloging librarians: Serving as re­ source persons for the authority control process; • Paraprofessional cataloging staff (li­ brary assistant III): Creating local author­ ity records for original cataloging mate­ rials and for series if these records are not available in the OCLC Authority File; • Paraprofessional cataloging staff (li­ brary assistants I and II): Searching, re­ viewing, and exporting authority records from the OCLC Authority File into the local system. 4. Substantial training for the catalog­ ing staff at library assistants I and II lev­ els is needed because of the complex na­ ture of the process. 5. The speed of cataloging new monographic titles is slowed down. The cataloging staff has to review authority headings at the time of cataloging. These additional steps on authority work have slowed the pace of processing new ma­ terials. � � Go. G o 404 College & Research Libraries September 2001 G G G S ;S 1 So ;S o oM S Soo Recommendations and Conclusion Based on the cataloging data, the im­ pact on cataloging work flow, and cost estimates, it was recommended to the library administration that in order to keep the local cataloging system consistent and current, two procedures must be in place (1) to validate headings for new biblio­ graphic records exported from the book vendor and from OCLC’s WorldCat on a regular basis, and (2) to update the existing authority in the local system whenever LC issues new headings that affect the local headings. There are 650,000 author­ ity records in the library’s catalog­ ing system. Each week, the LC issues a long list of headings, including both new headings and replace­ ments for old/obsolete headings. Because the cataloging staff has no way of knowing which existing heading in the local authority file needs to be replaced, the task of manually searching and comparing each heading issued weekly by LC will require a full-time cataloging staff member. After a thorough review of the current cataloging work flow, a bet­ ter understanding was gained that in-house authority work would take a large amount of staff time. It is almost impossible to keep the c a t a l o g i n g d a t a b a s e c u r r e n t through the in-house maintenance. Ample library literature has pro­ vided convincing data that the cost of outsourcing this procedure is lower than the cost of in-house maintenance. Therefore, the Tech­ n i c a l S e r v i c e s D i v i s i o n r e c o m ­ mended selecting a vendor to per­ form the authority control proce­ dures for the library. The library ad­ ministration is taking the recom­ mendations seriously. The funding for maintaining authority control is being included in the budget plans of the 2001/2002 fiscal year. Planning an Authority Control Project at a Medium-sized University Library 405 Notes 1. Robin K. Wendler, “Automating Heading Correction in a Large File: Harvard’s Experi­ ence,” in The Future Is Now: Reconciling Change and Continuity in Authority Control (Columbus, Ohio: OCLC, 1995), 5–10. 2. Henriette D. Avram, “Authority Control and Its Place,” Journal of Academic Librarianship 9 (Jan. 1984): 331–35. 3. Ibid., 333. 4. Larry Auld, “Authority Control: An Eighty-year Review,” Library Resources & Technical Services 26 (Oct./Dec., 1982): 319–30. 5. Barbara Tillett, “Automated Authority Control: A Survey of Literature,” seminar paper (Los Angeles: Graduate School of Library and Information Science, Univ. of California, June 1982), with corrections and additions, Oct. 1982. 6. Arlene G. Taylor, “Research and Theoretical Considerations in Authority Control,” in Au­ thority Control in the Online Environment: Considerations and Practices, ed. Barbara B. Tillett (New York and London: Haworth, 1989), 29–56. 7. http://www.oclc.org/firstsearch/databases/ LibraryLit database, accessed in Mar. 2001. 8. Taylor, “Research and Theoretical Considerations in Authority Control,” 33. 9. Joan M. Bechtel, “An Authority Control Alternative for Small Colleges,” College & Re­ search Libraries 53 (Nov. 1992): 485–98. 10. Elaine Peterson and Bonnie Johnson, “Is Authority Updating Worth the Price?” Technicali­ ties 10, no. 5 (May 1990): 1+. 11. Kerrie Talmac, “Authority Control,” in Technical Services Today and Tomorrow, 2nd ed. (Englewood, Colorado: Libraries Unlimited, 1998), 129–39. 12. Susan L. Tsui and Carol F. Hinders, “Cost-Effectiveness and Benefits of Outsourcing Au­ thority Control,” Cataloging & Classification Quarterly 26, no.4 (1998): 43–61. 13. Library Technology, Inc., documents at http://www.LibraryTech.Com/index.html, ac­ cessed in Mar. 2001. 14. Tsui and Hinders, “Cost-Effectiveness and Benefits of Outsourcing Authority Control,” 59. 15. Ibid., 59. http://www.LibraryTech.Com/index.html http://www.oclc.org/firstsearch/databases