Dataset for: Alternative metrics, traditional problems? Assessing gendered dynamics in the altmetrics of political science

Dataset and replication files for "Alternative metrics, traditional problems? Assessing gendered dynamics in the altmetrics of political science", published in European Political Science (2023). More detailed information in the README file. Please refer to the methodology section of the published article and the altmetric.py command file annotations for details. The abstract of the published article is below: Altmetrics are an emerging form of bibliometric measurement that capture the online dimension of scholarly exchange. Against the backdrop of both a higher education landscape increasingly focused on quantifying research productivity and impact, as well as literature emphasising the need to address gender bias in the discipline, we consider whether and how altmetrics (re)produce gendered dynamics in political science. Using a novel dataset on the Altmetric Attention Scores (AAS) of political science research, we investigate two questions: Do AAS vary by gender? And how do AAS relate to gendered social media dynamics? We find that AAS reproduce gendered dynamics found in disciplinary publication and citation practices. For example, journal articles authored exclusively by female scholars score 27% lower on average than exclusively male-authored outputs. However, men are also more likely to write articles with an AAS of zero. These patterns are shaped by the presence of high-scoring male “superstars” whose research attracts much online attention. Complementing existing scholarship, we show that the AAS closely overlaps with virality dynamics on Twitter. We suggest that these gendered dynamics may be hidden behind the seemingly neutral, technical character of altmetrics, which is worrisome where they are used to evaluate scholarship.