Errors in Knowledge Organization

In practical librarianship and bibliographical work there is usually a strong sense about whether things are done "correctly" or not. This question is not, however, much visible in research in Library and Information Science (LIS). There may be several reasons why this is the case, including lack of specific knowledge and reluctance to criticize the work done in the field. On the philosophical level one might claim that descriptions are not wrong or true, just more or less purposeful in relation to different needs, why the concept of "error" is meaningless. In spite of this argument, there are many cases in which given descriptions cannot be considered anything but errors. If this was not the case, there would be no need hire professionals to do KO. Attar (2006) thus presents typical cataloging errors provided by student catalogers. Errors in KO may imply search failures. Tonta (1992) provides a review of this field.

Bade provides no statistical data about the extent of the problems. The evidence he cites is selected case studies, and we are left to wonder how pervasive the problem is. Nobody have, as far as I know, provided quantitative studies of different kind of errors in databases. Chan & Vizine-Goetz (1997) report errors in subject heading assignment, but these focus mainly on whether assigned headings match current LC authorities, whether they are constructed properly, and so on, not on whether they accurately reflect the content of the work in question. Although statistical data is missing experienced librarians who regularly use these databases know that the problem is real. Bade makes it clear, however, after these errors are created in the bibliographic utilities, they propagate through local catalogs as librarians and library database managers too often blithely load records without adequate quality control at the local level.

Bade’s real concern is with linguistic errors, which he understands as cataloging errors that result from an inadequate understanding of the language in which the work is written, and intellectual errors, which he understands as errors that occur because of a lack of subject knowledge on the part of the cataloger. He recommends recognizing the intellectual nature of cataloging. One might whish it is possible to differentiate the concept of “intellectual errors” because the intellectual background for KO is what courses in information science and knowledge organization (including, of course domain specific courses) are designed for.

Bade (2007) finds that cataloguing in libraries was formerly done by cataloguers with a wide range of academic backgrounds and linguistic abilities. With the rise of networked databases much of this work is now automated, outsourced to vendors, or done by persons lacking the requisite skills. The removal of this activity from libraries leads to a generic product produced for a generic user, with no possibility for a library-internal evaluation of the product. Librarians demand "a bibliographic record" of a certain form in a manner analogous to the generative grammar's production of sentences. So long as the form is correct, it is not evaluated for appropriateness or usefulness. The resulting information is often equivalent to colorless green ideas.

"The chief defect of most of the literature on the correctness of bibliographical records has been that qualitative evaluation of these matters has been performed strictly on the basis of the records in the database – formal characteristics, completeness, conformity to standards – with no attempts to evaluate these records in terms of their appropriateness in respect of the actual item being described and their usefulness for the potential and future users. " (Bade, 2007, p. 65).

Drabenstott & Vizine-Goetz (1994) is about Library of Congress Subject Headings (LCSH) and their role in information retrieval. The book makes careful analysis of what kinds of queries a likely to succeed and what kinds of queries are not likely to succeed given LCSH.

Åström (2006) found that ". . .descriptors from LISA [Library and Information Science Abstracts] were hard to use due to inconsistencies and errors in indexing". Åström used instead the ERIC database without any analysis of the quality of indexing in this database.

Attar, K. E. (2006). Why appoint professionals? A student cataloguing project. Journal of Librarianship and Information Science, 38(3), 173-185.

Bade, D. W. (2002). The Creation and Persistence of Misinformation in Shared Library Catalogs: Language and Subject Knowledge in a Technological Era. Champaign-Urbana, Ill.: Graduate School of Library and Information Science, Univ. of Illinois (Occasional Papers, no. 211).

Bade, D. W. (2003). Misinformation and Meaning in Library Catalogs. Chicago: The author.

Bade, D. W. (2004). Theory and practice of bibliographic failure - or misinformation in the information society. Ulaanbaatar: Chuluunbat.

Bade, D. (2007). Colorless green ideals in the language of bibliographic description: Making sense and nonsense in libraries. Language & Communication, 27(1), 54-80.

Beall, J. (2006). Metadata and Data Quality Problems in the Digital Library. Journal of Digital Information, 6(3), article no. 355, 2005-06-12. http://jodi.tamu.edu/Articles/v06/i03/Beall/

Boserup, I. & Krarup, K. (1982). Reader-Oriented Indexing. An investigation into the extent to which subject specialists should be used for the indexing of documents by and for professional readers, based on a sample of sociological documents indexed with the help of the PRECIS indexing system. Copenhagen: The Royal Library.

Chan, L. M. & Vizine-Goetz, D. (1997). Errors and Obsolete Elements in Assigned Library of Congress Subject Headings: Implications for Subject Cataloging and Subject Authority Control. Library Resources & Technical Services, 41(4), 295-322.

Drabenstott, K. M. & Vizine-Goetz, D. (1994). Using Subject Headings for Online Retrieval: Theory, Practice, and Potential. San Diego: Academic Press.

Drabenstoff, K. M. & Weller, M. S. (1996). Failure analysis of subject searches in a test of a new design for subject access to online catalogs. Journal of the American Society for Information Science, 47(7), 519-537.

Lancaster, F. W. (2003). Indexing and Abstracting in Theory and Practice. 3. ed. London: Library Association Publishing.

Tonta, Y. (1992). Analysis of search failures in document retrieval systems: a review. The Public Access Computer Systems Review, 3(1), 4-53.