Changing predictive asymmetry: network agenda correspondence between state media and the user on China's fertility policy (2013-2025)

Over the past decade, China's fertility policy has shifted from restrictive population control toward pronatalist support, creating an important case for examining the adaptation of government communication in a platformized media environment.

We analyzed 126 fertility-policy posts published by People's Daily and 196,218 associated user comments on Weibo between 2013 and 2025. The LDA model was used to identify four topic-derived policy attributes. Event-based cross-lagged correlations evaluated directional associations against the Rozelle-Campbell baseline, while topic joint-intensity networks and event-level QAP correlations assessed aggregate structural correspondence. Analyses were conducted across three periods: the selective two-child reform period, the universal two-child policy period, and the three-child and pronatalist-support period.

Four topic-derived attributes emerged: Policy Change and Childrearing Burdens, Women's Employment and Work-Family Responsibilities, Gender Preference and Family Norms, and Maternity Care and Childbirth Support. In the first two periods, media-preceding-only patterns were descriptively more frequent than user-preceding-only patterns. In the third period, user-preceding-only patterns became more frequent. Gender analysis revealed gender- and period-specific differences in users' relative attention to the four policy attributes.

The relationship between the state-media and user agendas was heterogeneous rather than consistently media-led or user-led. Aggregate thematic similarity could coexist with weaker event-level structural correspondence. The findings are consistent with a potentially more reciprocal communication environment in the later policy period. It also indicates that fertility-policy communication should recognize gender-differentiated issue attention within the user agenda.