Exploring educational hypogamy among women in urban and rural China: Insights from random forest machine learning

Background: Educational hypogamy, where women marry men with lower educational attainment, reflects evolving gender roles and societal norms. In China, the rapid expansion of education, coupled with persistent traditional values, provides a unique context to study this phenomenon. Methods: Using data from the 2013, 2015, 2017, 2018, and 2021 waves of the China General Social Survey (CGSS), this study applies logistic regression models and Random Forest machine learning techniques to analyze the impact of education on women’s selection of hypogamy. Key control variables include age, income, parental education, and household registration, with a focus on urban-rural differences. Results: Between 2013 and 2021, the proportion of women choosing hypogamy increased from 17.42% to 20.06%. As education levels rose, so did the likelihood of choosing hypogamy, particularly among women with higher educational attainment. Other factors such as income and parental education displayed complex interactions with hypogamy, with urban women experiencing more nuanced influences compared to rural women, who showed a clearer education-driven pattern. The random forest analysis further confirmed education as the most significant predictor of hypogamy. Discussion/conclusion: The rise in educational hypogamy highlights women’s increasing autonomy and challenges to traditional gender norms, especially in rural areas where education’s impact is more pronounced. Urban-rural disparities suggest the need for targeted policies to promote gender equality. Future research should examine the long-term implications of educational hypogamy on household and child-rearing dynamics.