The quality of processing MARC21 item headers and coded content features in some national libraries

Lessons from an analysis

Abstract

In our previous work, we compared the processing quality of content data fields (648, 65X) in the MARC21 holdings of major national libraries. In this study, we focus on the processing level of mandatory item headers and of some coded content attributes (007 and 008). In these, sources are or should be classified according to an enumerative classification system consisting of one-character alphanumeric codes. It turned out that the use of positions in the latter two fields is extremely poor. This is related to their optional nature on the one hand, and to the shortcomings of the relevant MARC classification, as well as to the redundancies with other data fields (comment fields and content data fields) on the other, which lead to the neglect of the 00X fields. Their completion would, however, be extremely useful from the point of information retrieval – if searchers in library systems were trained to use user-friendly search options.