Search patterns of remote users: An analysis of OPAC transaction logs

Information Technology and Libraries

Volumn 12, Issue 3, 1993, Pages 321-339

  Millsap, Larry ^a   Ferl, Terry Ellen ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 21144479622     PISSN: 07309295     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (47)

1. Remote users issued 25% of the FIND commands during the period studied by the investigators. May 7-13, 1991, and 33% of the FIND commands during the week of October 17-23, 1992. During academic breaks and other periods of low instructional activity, remote use frequently rises to 50%.
2. Terry Ellen Ferl and Larry Millsap, "Remote Use of the University of California MELVYL Library System: An Online Survey," Information Technology and Libraries 11 (Sept. 1992): 285-303.
3. Statistics published on the system in November 1992 were 7,022,367 book titles and 792,980 periodical titles. Total volume holdings reported at that time were about thirteen mtflion. In addition to holdings of the UC libraries, the MELVYL system also reflects holdings of the California State Library, the California Academy of Sciences, and the Center for Research Libraries (Chicago).
4. Charles R. Hildreth, Intelligent Interfaces and Retrieval Methods for Subject Searching in Bibliographic Retrieval Systems (Washington, D.C.: Cataloging Distribution Service, Library of Congress, 1989), 6.
5. The MELVYL System Reference Manual (Oakland, Calif.: Division of Library Automation, Univ. of California, cl990), which is updated periodically, provides detailed information about the system, its databases, and patron interface.
6. Reproduced, with permission, from Search Sheets for OPACs on the Internet (Westport, Conn.: Meckler, 1991), 33-34.
7. Hildreth, Intelligent Interfaces, 2.
8. For an extensive review of current research on OPAC user interfaces, see Martha M. Yee, "System Design and Cataloging Meet the User: User Interfaces to Online Public Access Catalogs," Journal of the American Society for Information Science 42 (Mar. 1991): 78-98.
9. Clifford A. Lynch, "The Next Generation of Public Access Information Retrieval Systems for Research Libraries: Lessons from Ten Years of the MELVYL System," Information Technology and Libraries 11 (Dec. 1992): 414.
10. Charles R. Hildreth, "Beyond Boolean: Designing the Next Generation of Online Catalogs," Library Trends 35 (Spring 1987): 652.
11. RhondaN. Hunter, "SuccessesandFailuresof Patrons Searching the Online Catalog at a Large Academic Library: A Transaction Log Analysis," RQ 30 (Spring 1991): 395-402;
12. Steven D. Zinlc, "Monitoring User Search Success through Transaction Log Analysis: The WolfPAC Example," Reference Services Review 19 (Spring 1991): 49-56;
13. Thomas A. Peters, "When Smart People Fail: An Analysis of the Transaction Log of an Online Public Access CataIog,"Journal of"Academic Librarian-ship 15 (Nov. 1989): 267-73;
14. Sally W. Kalin The Searching Behavior of Remote Users: A Study of One Online Public Access Catalog (OPAC)" in Proceedings of the 54th Annual Meeting of the American Society for Information Science 28 1991 178-185
15. Kalin's project aim was "to determine if remote users of LIAS [Pennsylvania State University's online catalog] search any differently than those who search it in-house. Analyzing the searches would identify areas that might be problematic for remote users, and assist the Libraries in developing appropriate support services" (p. 178).
16. John E. Tolle, Current Utilization of Online Catalogs: A Transaction Log Analysis, vol.1 (Dublin, Ohio: OCLC Office of Research, 1983), 107-110
17. Lois Ann Colaianni, "Evaluating Online Catalogs: The Need for Data," in The Impact of Online Catalogs (New York: Neal-Schuman, 1986), 81-87.
18. Ferl and Millsap, "Remote Use," 288-289
19. For a description of the MELVYL transaction logging system, see Ray R. Larson s, "Transaction Logs: Gathering Information about MELVYL," DLA Bulletin 7 (Oct. 1984): 17-18;
20. and Ray R. Larson and Vicki Graham, "Monitoring and Evaluating MELVYL," In-formation Technology and Libraries 2 (Mar. 1983): 93-104. For the present study, the UC Division of Library Automation provided the investigators with detailed documents that described the transaction monitoring system and defined the various elements captured in the logs.
21. A zero retrieval in itself is neutral. It must always be viewed in some context to gain significance. For example, a knowledgeable observer may detect that a search word is misspelled, a command word is invalid, a search term does not match the controlled vocabulary of the index being searched, and so forth. Or the observer may note that a succession of searches strongly suggests the user is in the wrong database. Nevertheless, OPAC system designers and researchers rconcerned about large numbers of searches with zero retrievals. In 1988, Clifford Lynch stated that the statistics for zero retrievals in the MELVYL catalog were "alarming." At that time, about 31.5% of MELVYL searches in COMMAND mode resulted in zero retrievals. The figure still rat that level. For Lynch's concerns on this and related matters, see his "Large Database and Multiple Database Problems in Online Catalogs," in OPACs and Beyond (Dublin, Ohio: OCLC, 1989).
22. A MELVYL catalog session will "timeout" (end automatically) after ten to twelve minutes with no commands or actions from a user. This default setting maybe changed by a user for any amount of time between five and thirty minutes.
23. Only 1% of the surveyed users responded that they had received help from library staff in conducting their searches.
24. Ferl and Millsap, "Remote Use," 286.
25. The system logging programs, of course, capture total numbers in a myriad of categories. The Division of Library Automation provided the investigators with tables of remote user statistics captured during the survey period, some of which were more detailed than those routinely published on the system. These data occasionally helped inform the conclusions of the investigators.
26. The status profiles of the 292 users in the excluded group mirror those of the group whose logs were studied.
27. Users who chose the menu mode, called LOOKUP, rather than the COMMAND mode, did not have to type the command word FIND, but did have to select and type an index name followed by a search term or terms.
28. The user may modify, replace, or delete characters in a command that was just executed. This capability, however, involves three edit-ing steps, so it is essentially unused by the systems clientele. For a description of the feature, log on to the system and type EXPLAIN MODIFY.
29. The MELVYL catalog permits a BROWSE command in several indexes. Terms are processed as keywords, and the feature may help the user discover authoritative forms of names and subjects.
30. Chandra Prabha, "Managing Large Retrievals: A Problem of the 1990s?" in OPACs and Beyond (Dublin, Ohio: OCLC, 1989), 33.
31. Michael S. Berger, The MELVYL System: The Next Five Years and Beyond," Information Technology and Libraries 11 (June 1992): 155.
32. Joseph R. Matthews, Gary S. Lawrence, and Douglas K. Ferguson, eds., Using Online Catalogs: A Nationwide Survey (New York: Neal-Schuman, 1983), 177.
33. When the investigator tried searching in several different ways by authorized LC subject terms-"Grading and marking (Students)," "Higher education," "College students, Rating of-she encountered similar problems of irrelevant retrievals, too many retrievals, zero retrievals, or long searches.
34. Ray R. Larson, "The Decline of Subject Searching: Long-Term Trends and Patterns of Index Use in an Online Catalog," Journal of the American Society for Information Science 42 (Apr. 1991): 197-215.
35. Mike Berger, "The Patron Meets the MELVYL Catalog: A Short History of the MELVYL/Patron Interface,"DLA Bulletin 26 (Spring 1992): 7.
36. In the online questionnaire, the users were not asked whether they had "found what they wanted" or were "satisfied with their session." A yes or no answer to such questions would have added little to the analysis of the logs themselves. Instead, survey questions about help received and anticipated need for future help were asked in order to elicit from the user a general impression about ease of system use.
37. During the survey period, 1.9% of all remote user FIND commands in the MELVYL catalog contained the AT Iimiter to confine the search to a particular library, and 1.0% contained Boolean operands (AND, OR, AND NOT). In the same period, the figures for in-Iibrary users were 3.8% and 0.7%, respectively.
38. Hildreth, Intelligent Interfaces, 10-11.
39. Hildreth, "Beyond Boolean," 653.
40. As the source for Hjerppe s comment, Hildreth cites Hjerppes "Project HYPERCATalog: Visions and Preliminary Conceptions of an Extended and Enhanced Catalog" (Paper presented at IRFIS 6 Conference [International Research Forum in Information Science] in Fraseati, Italy. 15-18 Sept. 1985).
41. Lynch, "The Next Generation," 409-411
42. Michael Buckland has described several experimental "amenities" that would give the OPAC user a summary analysis of the composition of the retrieved set to help him decide what to do next. Buckland's contention that such amenities are needed and desirable is based in part on unpublished analyses of transaction logs that reveal "unexpectedly low levels of effectiveness in use of the MELVYL system (and probably of online catalogs generally)." See his "Agenda for Online Catalog Designers," Information Technology and Libraries 11 (June 1992): 157-163
43. One example of a heuristic feature in the MELVYL catalog is the following: If the user has zero results from a subject search, an onscreen message suggests that he may get results by reentering his search as a title word search. If the user also invokes the HELP screen offered at that point, additional strategies are suggested.
44. Lynch, "The Next Generation," 410.
45. Findings in a recent study by Bryce Allen and Gillian Allen suggest "that librarians should be careful to distinguish between their own abilities and those of their patrons when designing information systems, bibliographic instruction programs, and similar library services. The abilities of professionals may lead to preferences for sophisticated interfaces, complex instructional programs, or services that assume high levels of cognitive abilities, while their library patrons may find browse interfaces, simple point-of-need instruction, and basic services to be more appropriate to their levels of cognitive abilities."
46. See their "Cognitive Abilities of Academic Librarians and Their Patrons," College & Research Libraries 54 (Jan. 1993): 67-72.
47. Hunter, "Successes and Failures of Patrons," 402.