Assessing scientific output through self-evaluation: Evidence from four social science fields

• We asked social science researchers from the top 500 universities to list their top three publications. • A total of 2331 researchers from 42 countries responded to the survey. • Our study links researchers’ self-evaluated top publications with their detailed bibliometric profiles. • Most top-three publications that are books or in non-English languages are not indexed in Scopus. • Publications with stronger bibliometric statistics are more likely to be selected among the top three. We asked academics from the economics, political science, psychology, and sociology departments of the top 500 universities to state their top 3 publications. 2331 researchers who work in 42 countries responded to the survey. We collected the respondents’ Google Scholar (GS) and Scopus profiles to identify the publications which they did and did not select as their top 3 publications. Therefore, our study links researchers’ self-evaluated top publications with detailed bibliometric profiles, enabling a comparison between subjective assessments and bibliometric statistics. Around 30 % of respondents’ top 3 publications in political science and sociology are not indexed in Scopus, largely because many are books or non-English works. The top 3 publications demonstrate greater citation performance and are published in higher-impact journals compared to the average output. However, only 40 % of the publications identified as top 3 in self-evaluations also rank among the top 3 by citation performance. Self-evaluations offer additional insights that complement bibliometric measures, helping to address issues such as limited coverage, neglect of locally relevant research, and underrepresentation of certain academic fields.