Search Results
Title: Unobserved Family Effects on the Risk of a First Premarital Birth
Resulting in 1 citation.
| 1. |
Powers, Daniel A. |
|
Unobserved Family Effects on the Risk of a First Premarital Birth Social Science Research 30,1 (March 2001): 1-24. Also: http://www.sciencedirect.com/science/article/pii/S0049089X00906823 Cohort(s): NLSY79 Publisher: Academic Press, Inc. Keyword(s): Childbearing, Premarital/Nonmarital; Ethnic Differences; Fertility; First Birth; Peers/Peer influence/Peer relations; Racial Differences; Siblings Using data from the National Survey of Youth, proportional hazards models for clustered data are used to account for shared unobserved family-level traits (or frailty) associated with non-marital childbearing. The variance in frailty can be used to evaluate a woman's conditional hazard of a first premarital birth if one or more of her sisters were to experience a premarital birth relative to her risk if none of her sisters experience a premarital birth. I find that among non-Hispanic White women, a first premarital birth by one sister doubles our estimate of another sister's risk of a first premarital birth after controlling for observed family-level & individual-level characteristics. Significant associations exist between several socioeconomic measures & the estimated frailty among White families. For a Black woman in the NLSY, the estimate of the risk of a first premarital birth would increase by only 14% if one of her sisters were to experience a first premarital birth. Low variance in frailty among Black families may be a result of the high prevalence of non-marital births in Black communities. Whether unobserved neighborhood, community, or peer-group traits contribute more to a Black woman's risk than unobserved family-level traits remains an important question for further research. 10 Tables, 1 Appendix, 41 References. [Copyright 2001 Academic Press.] |
|
|
Bibliography Citation
Powers, Daniel A. "Unobserved Family Effects on the Risk of a First Premarital Birth." Social Science Research 30,1 (March 2001): 1-24.
|