Why discrepancies in searching the conservation biology literature matter

Conservation biologists seek as much information as possible for evidence-based conservation actions, so they have a special concern for variations in literature retrieval. We assessed the significance for biological conservation of differences in literature retrieval across databases by comparing five simple subject searches in Scopus, Web of Science (WoS) (comparing two different subscriptions), Web of Science (Core Collection) (WosCC) (comparing two different subscriptions) and Google Scholar (GS). The efficiency of a search (the number of references retrieved by a database as a percentage of the total number retrieved across all databases) ranged from 5% to 92%. Different subscriptions to WoS and WoSCC returned different numbers of references. Additionally, we asked 114 conservation biologists which databases they used, their awareness of differing search options within databases and their awareness of different subscription options. The four most widely used databases were GS (88%), WoS (59%), WoSCC (58%) and Scopus (27%). Most respondents (≥ 65%) were unsure about specific features in databases, although 66% knew of the service GS Citations, and 76% agreed that GS retrieved grey literature effectively. Respondents' publication history did not influence their responses. Researchers seeking comprehensive literature reviews should consult multiple databases, with online searches using GS important for locating books, book chapters and grey literature. Comparative evaluations of publication outputs of researchers or departments are susceptible to variations in content between databases and different subscriptions of the same database, so researchers should justify the databases used and, if applicable, the subscriptions. Students value convenience over thoroughness in literature searches, so relevant education is needed.