学历性别反超背景下的就业行业推荐模型研究——基于性别、学历、专业与学校多因子匹配的分析

Research on an Employment Sector Recommendation Model against the Backdrop of Women Surpassing Men in Educational Attainment —An Analysis Based on Multi-Factor Matching of Gender, Educational Background, Major and School

摘要: 随着政府在教育公平工作上的重视，人们对高等教育的追求有更深的向往，逐渐地，社会不再以“男性”为主导，女性有了更多的机会享受教育，一些活动和工作岗位也特意为女性而展开。于是，女性在教育体系中的参与度逐渐上升，呈现出了“学历反超”的现象。但当前就业市场仍体现出显著的性别分化。根据2023年《中国劳动统计年鉴》，技术类岗位男性占比达76%，而25~29岁女性本科比例已反超男性30%。在体制内，公务员招录中偏好男性的岗位比例高达22.36%，而明确偏好女性的岗位几乎为零。因此，本文采用第五次(2000)、第七次(2020)人口普查数据，通过Excel数据预处理分析了这两个时间段展现出的男女学历变化趋势，并总结了各行各业的男女比例。再基于社会现象，借助Tableau可视化平台，构建四因子推荐模型，其中包含学历匹配度、专业匹配度、学校等级匹配度与性别友好评分四项指标，开发出交互式就业推荐仪表盘，为高校毕业生提供就业选择平台。并建议根据实时环境，增添多关联因子，推荐至社会中，使招聘标准透明化、合理化。

Abstract: As the government places increasing emphasis on educational equity, public aspiration for higher education has grown stronger. Gradually, society has moved away from male dominance, and women have gained greater access to educational opportunities. Specific initiatives and job positions have also been created explicitly for women. As a result, female participation in the education system has continued to rise, giving rise to the phenomenon of “academic reversal”, where women have surpassed men in educational attainment. However, the current job market still exhibits significant gender segregation. According to the 2023 China Labour Statistical Yearbook, men occupy 76% of technical positions, while the proportion of women aged 25~29 with bachelor’s degrees has exceeded that of men by 30 percentage points. Within the public sector, 22.36% of civil service recruitment positions show a preference for male candidates, whereas almost none explicitly favour females. Therefore, this study utilizes data from the Fifth (2000) and Seventh (2020) National Population Censuses. Through data preprocessing in Excel, we analyse the trends in educational attainment by gender across these two periods and summarize gender ratios across various industries. Building on these social observations, we employ the Tableau visualization platform to construct a four-factor recommendation model. This model includes four indicators: educational background match, major match, institutional prestige match, and a gender-friendliness score. An interactive employment recommendation dashboard has been developed to provide a platform for university graduates to explore career options. Furthermore, we propose incorporating additional relevant factors in real time to extend the model’s application in society, thereby promoting more transparent and equitable recruitment standards.