Note: This page was written from training-data knowledge; the raw PDF (ProQuest scan) could not be extracted. Claims should be verified against the original before being cited.
Published in American Sociological Review 61(3): 386–406 (1996). Wu uses discrete-time event-history models on the National Longitudinal Survey of Youth (NLSY) to study what raises young women's risk of a first premarital birth. The paper's central contribution is distinguishing three distinct predictors — family structure instability during childhood, the level of family income, and the instability (variability) of that income — and showing that all three operate independently. The key novel finding is that income instability significantly elevates premarital birth risk even after controlling for income level, suggesting that income volatility itself — not just poverty — is a mechanism through which economic disadvantage translates into early nonmarital childbearing.
(Source text unavailable — ProQuest scan not extractable.)
The income-instability finding is this paper's lasting contribution. Prior work emphasized income level; Wu shows that volatility — income that swings year to year — independently raises nonmarital birth risk. This anticipates later work on precarity and family formation. The paper is also methodologically important as an early application of discrete-time hazard models with time-varying economic covariates to fertility outcomes using the NLSY. The limitation is the observational design: selection into low-income, unstable households confounds the income effects, and the paper relies on statistical controls rather than causal identification. By contemporary standards the causal identification is weak; the paper should be read as establishing correlational patterns and motivating mechanisms rather than estimating causal effects.