Revision of author abstracts: how it is carried out by LISA editors

The literature on abstracts recommends the revision of author supplied abstracts before their inclusion in database collections. However, little guidance is given on how to carry out such revision, and few studies exist on this topic. The purpose of this research paper is to first survey 187 bibliographic databases to ascertain how many did revise abstracts, and then study the practical amendments made by one of these, i.e. LISA (Library and Information Science Abstracts).

Database policies were established by e‐mail or through alternative sources, with 136 databases out of 187 exhaustively documented. Differences between 100 author‐supplied abstracts and the corresponding 100 LISA amended abstracts were classified into sentence‐level and beyond sentence‐level categories, and then as additions, deletions and rephrasing of text.

Revision of author abstracts was carried out by 66 databases, but in just 32 cases did it imply more than spelling, shortening of length and formula representation. In LISA, amendments were often non‐systematic and inconsistent, but still pointed to significant aspects which were discussed.

Amendments made by LISA editors are important in multi‐ and inter‐disciplinary research, since they tend to clarify certain aspects such as terminology, and suggest that abstracts should not always be considered as substitutes for the original document. From this point‐of‐view, the revision of abstracts can be considered as an important factor in enhancing a database's quality.