Database source coverage: hypes, vital signs and reality checks

The purpose of this paper is to show that the extent of journal coverage, size and retrospectivity are three of the most commonly advertised features of database content. However, these data by themselves rarely provide a good enough sense of the real breadth of database coverage.

Some easy‐to‐calculate measures can provide vital signs of database coverage, which in turn may trigger in‐depth analysis at the individual journal level. When similar vital signs are produced from comparable databases and database segments, as is done for Arts and Humanities (A&H) sources in Web of Science, Scopus, H.W. Wilson and CSA databases for this paper, the real dimensions of the source coverage can be much better understood.

The decision to almost double the number of A&H journals reinforces the concept of enhancing the breadth of source coverage, obviously at the expense of depth of coverage. It is like those “If it is Tuesday, it must be Belgium” tours, where the emphasis is on beating the other tours by advertising more comprehensive coverage of European countries by “virtue” of visiting more countries in a single afternoon.

For scientometric searches so called “known item” searches are made, and the more records that are missing from core journals, the more harm is done to the subjects of the search. In subject searches a few good finds may be fine, but in “known item” searches that is not good enough. Doubling the cited reference enhanced segment, and eliminating the gaps in, say, the 1975‐2009 segment would have been much smarter policy for Scopus.