Evidence Based Library and Information Practice Evidence Based Library and Information Practice 2010, 5.4 115 Evidence Based Library and Information Practice Commentary Evidence Based Cataloguing: Moving Beyond the Rules Kathy Carter Projects Librarian, Bibliographic and Information Technology Services University of Alberta Libraries Edmonton, Alberta, Canada Email: kathy.carter@ualberta.ca Received: 11 Oct. 2010 Accepted: 18 Oct. 2010 2010 Carter. This is an Open Access article distributed under the terms of the Creative Commons-Attribution- Noncommercial-Share Alike License 2.5 Canada (http://creativecommons.org/licenses/by-nc-sa/2.5/ca/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one. Cataloguing is sometimes regarded as a rule- bound, production-based activity that offers little scope for professional judgement and decision-making. In reality, cataloguing involves challenging decisions that can have significant service and financial impacts. The current environment for cataloguing is a maelstrom of changing demands and competing visions for the future. With information-seekers turning en masse to Google and their behaviour receiving greater attention, library vendors are offering “discovery layer” products to replace traditional OPACs, and cataloguers are examining and debating a transformed version of their descriptive cataloguing rules (Resource Description and Access or RDA). In his “Perceptions of the future of cataloging: Is the sky really falling?” (2009), Ivey provides a good summary of this environment. At the same time, myriad new metadata formats and schema are being developed and applied for digital collections in libraries and other institutions. In today’s libraries, cataloguing is no longer limited to management of traditional AACR and MARC-based metadata for traditional library collections. And like their parent institutions, libraries cannot ignore growing pressures to demonstrate accountability and tangible value provided by their services. More than ever, research and an evidence based approach can help guide cataloguing decision-making. Decisions Librarians face a variety of cataloguing-related decisions that arise in various settings. These are only a few of the issues that are challenging us at the present time. At the level of international standards and practice, • Can one set of descriptive rules serve for all types of information objects? • What metadata are required to meet the differing needs of different users? • Will RDA produce records that satisfy user needs? • When can MARC be abandoned and what should replace it? mailto:kathy.carter@ualberta.ca� Evidence Based Library and Information Practice 2010, 5.4 116 • How can libraries share metadata design and creation with other players such as publishers, indexers, archives, and authors? At the level of individual libraries, for traditional cataloguing, • Which sources of metadata records are the most comprehensive, the best quality, the cheapest, the most suited to the library’s collection and policies? Possibilities include OCLC, book and other library vendors, SkyRiver, and other library catalogues. • What sort of record checking, upgrading and customization adds tangible value? • Can savings be realized by outsourcing some work, such as copy cataloguing, original metadata creation, or authority control? And as other forms of metadata and discovery are added to traditional cataloguing, • How and when can dissimilar records be combined effectively in a single database or search? • What are the needs of our different user communities and how can we differentiate our metadata content, retrieval and display for each? • Can a Virtual International Authority File (VIAF) improve author disambiguation across multiple metadata sources, or would programmatic disambiguation solutions be more effective? • How important is consistency of data such as author, genre and geographic headings? • What software and system capabilities are most important for efficient, scalable operations and services? • What are the most effective ways to expose our metadata and integrate it into the web tools and venues where our users are? Metadata Over time, metadata is subject to change and reuse; it may be enhanced, corrected, indexed, extracted, merged, converted, crosswalked, and reformatted. But most importantly, metadata endures. WorldCat contains records created centuries ago, cheek by jowl with millions of others created since, right up until just seconds ago. Of equal importance, metadata is widely shared, reflecting the goal of minimizing redundant effort. Decisions regarding metadata content and standards are made in this context of long-term and shared record use in aggregated environments. Typically in the traditional cataloguing context, standards are developed by national and international bodies, changes are not made lightly or quickly, and practitioners tend to develop a habitual reliance on rules rather than evidence. Conversely, creators of metadata for digital collections such as institutional repositories have often shunned traditional cataloguing structures such as AACR and MARC, and developed standards that are better suited to individual collections and to networked resources and services. RDA attempts to serve both these worlds, but has been the object of vigorous criticism from both.i Management of Cataloguing Operations The organization, staffing, and supervision of cataloguing operations, and management of purchased or locally developed services and systems, are significant issues and cost centers for libraries. Some of the issues are similar to management issues in other library and non- library settings, and they are more amenable than metadata standards to local decision- making. Yet operational decision-making cannot be separated from definition and assessment of desired outcomes. The difficulties in achieving this can be seen in the Final report of the Task Force on Cost/Value Assessment of Bibliographic Control of the ALA Heads of Technical Services in Large Research Libraries Interest Group (2010). Charged with “identifying measures of the cost, benefit, and value of bibliographic control for key stakeholder communities, and developing a plan for implementing these measures” (p.2), the Task Group noted that: Evidence Based Library and Information Practice 2010, 5.4 117 The objective of this work was… to begin to identify sound measures that can inform decisions by those engaged in the creation, exchange, and use of bibliographic data. Our ability to make sound decisions and mindful changes around bibliographic control is hindered by our lack of operational definitions of value and methodologies for assessing value within our institutions. (p.2) Cataloguing Research Some aspects of cataloguing, such as the effectiveness of keywords vs. controlled vocabularies for subject searching, have been subject to detailed research.ii Yet the scope, complexity, and uses of cataloguing present tremendous challenges to designing and conducting relevant research. In 1997 Ling Hwey Jeng observed “the lack of empirical studies of many fundamental theories in cataloging” (p.124), and in 2008 Janet Swan Hill noted “a persistent shortcoming in the decision-making process that needs to be addressed is the lack of serious research into user needs and benefits, and the actual impact on users of database quality decisions” (p.5). In recent years OCLC has gone some distance to reduce this shortcoming, with research into metadata content and user behaviour and expectations. In their 2009 study “Online catalogs: What users and librarians want” (OCLC) provides a welcome well-grounded insight into users’ priorities, and a revealing comparison with how they differ from those of librarians. Yet it is discouraging that the current extensive testing of RDA has as its goal “to assure the operational, technical, and economic feasibility of RDA” (Library of Congress, 2010, para. 1) with no attempt to include tests of its effectiveness in serving users. Inspired by On the Record: Report of The Library of Congress Working Group on the Future of Bibliographic Control (2008), the American Library Association’s Association for Library Collections & Technical Services (ALCTS) declared 2010 as the Year of Cataloging Research (News from ALCTS, 2009). This encouraging development should help build awareness of the value of research. Barriers to Evidence Based Practice in Cataloguing In addition to the paucity of relevant cataloguing-related research, it can be difficult to find results of research that has been conducted. Organisations and vendors who carry out and analyze research in support of their own product development often do not release it publicly, and research done by libraries to guide their own decisions is sometimes not publicised or distributed beyond their own institutions. Relevant research, especially related to user interfaces and user behaviour, is done in non-library disciplines and may be overlooked. Other barriers that can result in failure to seek and apply research findings to decision-making are lack of time or management support and a perceived disconnect between researchers and practitioners. In addition, the pace and variety of new technology and application developments creates pressure to implement novel services quickly and makes their assessment challenging. In cataloguing, there are powerful pressures to accept rules, records, and systems created by others. The dependencies resulting from this mode of practice can undermine cataloguers’ confidence in their own judgement and their ability to influence decisions about catalogue design and other uses of metadata. Adopting an evidence based approach is one way to counteract and break out of this marginalization. Learning from Experience In 2006 the Library of Congress (LC) made a decision to cease creating series authority records and providing controlled series access points. Many aspects of this decision and its ramifications do not offer a good model of evidence based practice. In announcing the change, LC representatives cited adequate series access provided by more powerful Evidence Based Library and Information Practice 2010, 5.4 118 indexing and keyword access (“LC to cease providing,” 2006), and an analysis that indicated “a substantial amount of cataloging time could be saved in the area of series control” (CONSER, 2006). These reasons may have been valid, but no evidence for them was provided at the time or subsequently. In response to the change, OCLC adjusted its record loading procedures to ensure that controlled series access would not be lost through overlay by an LC record, and the Program for Cooperative Cataloging (PCC) created guidelines for member libraries who decided to continue series authority control in their records. Both OCLC and PCC allowed members the option of providing series control or not in their records. To the author’s knowledge, no further investigation has been done on the effects of these decisions. Data about the following questions would provide valuable evidence for decision-making by individual libraries and national agencies. • What operational savings were realized by LC? • What were the costs for OCLC in changing its record loading procedures and subsequent record maintenance? • What are the additional costs for libraries that continued to control series headings and maintain series authority records? • What was the impact, if any, on the operations of libraries that adopted LC’s policy? • What was the impact, if any, on catalogue users? • What is the cost difference between distributed series authority control and centralised series control by LC and other national libraries? Would the same difference apply to authority control of other access points such as authors? Development of the CONSER Standard Record for serials in 2007 demonstrates a more successful application of several principles of evidencebased practice. • Objectives were clearly defined (Functionality, Cost-Effectiveness, and Conformity to current standards) • Testing was carried out in 14 libraries to assess the success of proposed guidelines in meeting the objectives. Results provided evidence of improved functionality and cost- effectiveness. • Changes to the Standard Record were made as a result of test findings. • Methodologies and findings were well documented and disseminated (Access Level Record for Serials Working Group, 2006). • Further testing is being done after implementation of the guidelines (Terrill, 2009). With this information, a decision on whether or not to adopt the CONSER Standard Record is much easier and more defensible. Using an Evidence Based Approach in Cataloguing Research and evidence cannot provide all the answers for the difficult decisions faced by cataloguers, but our professional judgement and accountability are strengthened by a critical and evidence based approach in our practice. Despite the many barriers, we can develop the habit of using research and evidence in everyday decision-making, and by example encourage colleagues to do likewise. Here are some suggested ways to use and support evidence based cataloguing: • Follow discussion lists, blogs, etc. to find out about articles, conference presentations and reports that describe research and findings. Contact authors for further data if appropriate. Don’t limit yourself to research by and for librarians. Especially relevant is the growing body of information on information-seeking behaviour. iii • Many decisions about metadata are appropriately made at the national and international level. As a practitioner, Be informed. Evidence Based Library and Information Practice 2010, 5.4 119 provide input to bodies that develop and decide on metadata standards, and encourage evidence based approaches in their work. If possible, volunteer to serve. Participate in testing of proposed new rules and standards, and share methodologies and findings. Examples are the test of the BIBCO Standard Record devised, carried out and reported by cataloguers at the University of Washington Libraries (Brooking, 2010) and the RDA testing currently underway in many libraries. Be involved. • As a practitioner, collaborate with researchers in designing and conducting relevant research. Use available data such as statistics from your ILS and OPAC. Replicate research done by others in order to verify (or not) their findings. Follow up on their suggestions for further research and methodology improvements. Report on unsuccessful and unexpected methods and findings; these can be just as useful as evidence of successes. Disseminate your results so others can benefit from both your findings and your methods. This can be done in a variety of formal and informal ways, e.g. articles, conference presentations, posting reports on library and personal websites or blogs, or listserv discussions. Be vocal. • Use the opportunities provided by development or analysis of metadata schema for local digital collections to incorporate evidence based decisions. Be resourceful. • When making decisions such as local variations in metadata standards, assess them for tangible evidence of improved outcomes. Use research findings to demonstrate to others in your institution the value added by cataloguing. Let them know how your decisions have been influenced by evidence. Resist decision- making based on anecdote, assumption, and temporary expediency. Encourage and model evidence based practice as an approach, not a rigid process that inhibits innovation. Be persistent. The organization of information through cataloguing is a core activity of librarianship. At a time when the value and sustainability of library cataloguing is being seriously questioned within our profession, the pursuit and application of empirical evidence offer a way forward to a rejuvenated cataloguing culture and practice. References Access Level Record for Serials Working Group. (2006, July 24). Final report. Retrieved October 8, 2010, from http://www.loc.gov/acq/conser/alrFina lReport.html Brooking, D. M., Kiegel, J. A., & Schiff, A. L. (2010, June 14). Test of the BIBCO Standard Record. Retrieved October 8, 2010, from http://staffweb.lib.washington.edu/co mmittees/CPC/bsr Case, D. O. (2007). Looking for information: A survey of research on information seeking, needs and behavior (2nd ed.). Amsterdam: Elsevier/Academic Press. CONSER. (2006, April 27). Joint BIBCO/CONSER meeting. Retrieved October 8, 2010, from http://www.loc.gov/acq/conser/conop 2006.html Gross, T., & Taylor, A. G. (2005). What have we got to lose? The effect of controlled vocabulary on keyword searching results. College & Research Libraries , 66, 212-230. Hill, J. S. (2008). Is it worth it? Management issues related to database quality. Cataloging & Classification quarterly , 46 (1), 5-26. Ivey, R. T. (2009). Perceptions of the future of cataloging: Is the sky really falling? Cataloging & Classification Quarterly , 47 (5), 464-482. Evidence Based Library and Information Practice 2010, 5.4 120 Jeng, L. H. (1997). Knowledge, technology, and research in cataloging. Cataloging & Classification Quarterly , 24 (1/2), 113- 127. LC to cease providing controlled series access. (2006, April 21). Posted to AUTOCAT discussion list. Library of Congress. (n.d.). Testing Resource Description and Access (RDA). Retrieved Oct. 8, 2010, from www.loc.gov/bibliographic- future/rda/ Library of Congress Working Group on the Future of Bibliographic Control. (2008, January 9). On the record. Retrieved October 27, 2010, from http://www.loc.gov/bibliographic- future/news/lcwg-ontherecord-jan08- final.pdf News from ALCTS. (2009). ALCTS Newsletter Online, ANO , 20 (4). OCLC. (2009). Online catalogs: What users and librarians want. Retrieved October 8, 2010, from http://www.oclc.org/reports/onlinecat alogs/fullreport.pdf Task Force on Cost/Value Assessment of Bibliographic Control, Heads of Technical Services of Large Research Libraries Interest Group, American Library Association. (2010). Final report. Retrieved October 8, 2010, from http://connect.ala.org/files/7981/costva luetaskforcereport2010_06_18_pdf_47 879.pdf Terrill, L. J. (2009). A snapshot of early acceptance of the CONSER Standard Record in local catalogs. Serials review , 35 (1), 16-27. Footnotes i See for example the archives of the NGC4lib discussion list (http://listserv.nd.edu/archives/ngc4lib.html) under subject “Cooperative Cataloging Rules Announcement” and of the AUTOCAT discussion list (https://listserv.syr.edu/scripts/wa.exe?A0=AUTOCAT) under subject “RDA and the Library Discovery Experience”. ii See for example T. Gross & Taylor, A.G. (2005) “What Have We Got to Lose? The Effect of Controlled Vocabulary on Keyword Searching Results” College & Research Libraries. 66, 212-230. iii See for example Case, D.O. (2007) Looking for information: a survey of research on information seeking, needs, and behavior (2nd ed.) Amsterdam: Elsevier/Academic Press. http://listserv.nd.edu/archives/ngc4lib.html� https://listserv.syr.edu/scripts/wa.exe?A0=AUTOCAT� / Evidence Based Library and Information Practice Commentary