中文摘要
我们研究了资深与初级研究人员之间的研究相似度在多大程度上与学术界的晋升相关，并考察了其对教职员工性别多样性的影响。利用意大利经济学终身教轨助理教授职位的全样本求职申请数据，并对数据集中学者每篇出版物的摘要应用自然语言处理技术（即文档嵌入），我们提出了一种新颖的研究相似度测度，该测度能够捕捉候选人与选拔委员会成员在研究主题、方法论或政策相关性上的贴近程度。我们表明，相似度与获胜概率强相关。此外，尽管在平均相似度上不存在性别差异，但女性候选人与选拔委员会成员的最大相似度更低。当仅聚焦女性委员会成员计算相似度时，该性别差距便消失了。结果表明，男性主导环境中的相似性偏误可能对性别与研究多样性产生影响。
Abstract
Abstract We investigate the extent to which research similarity between senior and junior researchers is related to promotion in academia and study implications for gender diversity among academic staff. Using data on the universe of job applications for tenure-track assistant professor positions in economics in Italy, and applying natural language processing techniques (i.e., document embeddings) to the abstract of each publication of the scholars in our dataset, we propose a novel measure of research similarity that can capture the closeness in research topics, methodologies or policy relevance between candidates and members of selection committees. We show that the degree of similarity is strongly associated with the probability of winning. Moreover, while there are no gender differences in mean similarity, the maximum similarity with selection committee members is lower for female candidates. This gender gap disappears when similarity is calculated by focusing only on female committee members. The results suggest that similarity bias in male-dominated environments may have implications for gender and research diversity.