Quantifying the sources of heterogeneity of fertility preferences in China

Rui Lu*, Gert Stulp, Anne H. Gauthier

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

15 Downloads (Pure)


There are no unanimous estimates on fertility preferences in China; they vary considerably across studies. Understanding this cross-study heterogeneity of reported preferred family size will contribute to the debates on the uncertainty of fertility in China. This paper quantified how much heterogeneity in aggregate-level fertility preferences can be explained by taking into account the demographic characteristics of the sample, its geographical location, and the measurement of fertility preferences. We retrieved 124 estimates of average preferred family size from 94 publications (from the year 2001 and onwards) in the China National Knowledge Infrastructure database (CNKI) and used mixed-effect meta-regression model to investigate heterogeneity. We found that the average preferred number of children in China since 2001 ranged from 0.91 to 2.88, with an average of 1.70 (SD = 0.26). Demographic differences accounted for 23%, human geographic regionalization for 21%, and the measurement of fertility preferences for 13% of the between-study variation. Sample characteristics (demographic and regional differences) and the measurement of fertility preferences together accounted for 46% of the variation in average preferred fertility. The averaged preferred family size varies by different characteristics and measurement. This study also points out whether particular groups have higher or lower preferences. It helps researchers to put estimates from fertility preferences from single studies in perspective. Moreover, this study can help resolve why researchers have such different outlooks on the future fertility of China.
Original languageEnglish
Pages (from-to)127-140
Number of pages14
JournalChina Population and Development Studies
Issue number2
Publication statusPublished - 2022

Cite this