Leukhina (2024) Study Population and Design

What the evidence says

Two PLoS ONE studies (2024) supply the population and design detail extracted for this topic, both drawn from large Chinese survey instruments and both restricted to fertility-relevant age bands. Sichen Liu (2024) draws on China's Migrant Population Dynamic Monitoring data, reporting a working sample of 110,667 married migrant families with one child, pooled from 2016 and 2018 survey waves; broken out by wave, the source states 60,764 eligible samples in 2016 and 43,167 eligible samples in 2018. The survey instrument behind that sample has broad geographic reach: 31 provinces and autonomous regions, 106 cities (spanning provincial capitals, planned cities, prefecture-level cities, and county-level cities), 2,456 streets, and 4,912 neighborhood committees or administrative villages. The study population was narrowed at the analysis stage to married individuals with one child, within the childbearing age range of 15 to 49 years. Its two explanatory variables were built from self-reported plans to have children in the next two years and self-reported current contraceptive use by the respondent and spouse.

Jun Li (2024) uses a separate survey with a different sampling architecture and a smaller analytic sample. After dropping records with missing values on selected variables, 13,203 individuals remained in the study. The age criteria differ by sex: women aged 20 to 49 and men aged 22 to 60 were retained. Geographically, the sample spans 153 counties across China. The underlying survey was built on a three-stage stratified probability sampling method with probability proportional to size (PPS): administrative districts or counties were selected first, then administrative villages or neighborhood committees within those, then households within those.

Neither source in the current set reports an estimated policy or behavioral effect — both claims sets describe sample construction, eligibility criteria, and survey architecture rather than measured outcomes. No claim here quantifies a relationship between any policy lever and a fertility, migration, or contraceptive outcome; the extraction covers methodology only.

Where the evidence splits

No contradictions in the current sources.

Thresholds and dose-response

No threshold established in the current sources.

Open empirical questions

• Neither source's claims here report an estimated effect of any policy or behavioral variable on a fertility, migration, or contraceptive outcome — what results these samples produced is not covered by the current extraction.
• Liu (2024)'s reported wave-level counts (60,764 in 2016 plus 43,167 in 2018) sum to fewer than the stated pooled total of 110,667; the source claims do not explain the difference.
• Whether findings from a migrant-specific sample (Liu) generalize to the broader, non-migrant population sampled by Li is not addressed by either source.
• Whether the differing age-eligibility rules (15–49 for both sexes in Liu; 20–49 women / 22–60 men in Li) meaningfully change comparability of any downstream estimates is not addressed by either source.