China's two-child policy, aimed at boosting the country's total fertility rate, has failed to achieve the desired outcomes. Previous studies on low fertility rates mainly used data obtained from demographic censuses, questionnaires, or interviews. These data-gathering methods are costly, entailing time delays and yielding limited information. User-generated content (UGC) provides an alternative data source. We propose a machine-human hybrid approach using UGC obtained from social media to assess users' intentions to have a second child. Our results showed that couples associate a second child with high economic costs mainly through negative impacts on the mothers' careers, with no concomitant economic benefits. A key motivation for having two children relates to the mental benefit of the joy in having children. However, raising a second child also entails considerable mental costs such as exhaustion and pressure. Couples largely seek help within their extended families, that is, their parents are major sources of child-rearing support. Therefore, the government should devise ways of reducing the negative impacts of having a second child on a woman's career and provide child-rearing support to help increase the fertility rate. Our proposed approach can also be used to elicit the reasons for low fertility rates in other countries.
Keywords: fertility intention; machine–human hybrid approach; second child; two-child policy; user-generated content.