Search Results

Author: Tuma, Nancy Brandon
Resulting in 3 citations.
1. Michael, Robert T.
Tuma, Nancy Brandon
Entry into Marriage and Parenthood by Young Adults
Demography 22,4 (November 1985): 515-544.
Also: http://www.springerlink.com/content/7m4gt2035165549w/
Cohort(s): NLSY79
Publisher: Population Association of America
Keyword(s): Adolescent Fertility; Family Background; Fertility; First Birth; Hispanics; Marital Status

Permission to reprint the abstract has not been received from the publisher.

This paper investigates the marital and fertility patterns of young men and women (age 14-21) using the first year data from the NLSY. The paper's substantive focuses are the influences of family background on early (teenage) entry into marriage and parenthood and the extent to which measured family characteristics can explain the large differences among whites, Hispanics and blacks. The paper's methodological focus is the comparison of results when a data set is analyzed as either a conventional cross-section file using a linear probability (OLS regression) or logistic (maximum likelihood) model or as a continuous time, event-history file using a partial likelihood model.
Bibliography Citation
Michael, Robert T. and Nancy Brandon Tuma. "Entry into Marriage and Parenthood by Young Adults." Demography 22,4 (November 1985): 515-544.
2. Michael, Robert T.
Tuma, Nancy Brandon
Youth Employment: Does Life Begin at 16?
Journal of Labor Economics 2,4 (October 1984): 464-476.
Also: http://www.jstor.org/stable/2534809
Cohort(s): NLSY79
Publisher: University of Chicago Press
Keyword(s): Employment, In-School; Employment, Youth; Racial Differences; Teenagers

Theoretical economic models, official labor force statistics, and most empirical studies of young workers disregard employment experience of students under age 16. Evidence from several sources, however, suggests that students ages 14 and 15 acquire substantial employment experience. Moreover, that experience is vastly different for black and white youths. Several policy-related issues, including causes of black-white differences in adult earnings, may deserve to be interpreted differently in the light of differentials in early employment experience. This employment experience of 14- and 15-year-olds in general and its racial pattern in particular should not continue to be ignored.
Bibliography Citation
Michael, Robert T. and Nancy Brandon Tuma. "Youth Employment: Does Life Begin at 16?" Journal of Labor Economics 2,4 (October 1984): 464-476.
3. Tuma, Nancy Brandon
Michael, Robert T.
A Comparison of Statistical Models for Life Course Analysis with an Application to First Marriage
In: Current Perspectives on Aging and the Life Cycle, Volume 2, Family Relations in Life Course Perspective. Z. Blau, ed. Greenwich, CT: JAI Press, 1986: pp. 107-146
Cohort(s): NLSY79
Publisher: JAI Press, Inc.
Keyword(s): Family Background; Life Course; Life Cycle Research; Marriage; Modeling, Logit; Modeling, Mixed Effects; Variables, Independent - Covariate

This paper addresses the question: "How similar are results pertaining to the effects of family background on early marriage when five different statistical models are used to analyze the data?" Data from the NLSY are used in this study of first marriage rates up to age 22 using three proportional hazard models--a Gompertz model, a Cox model, and a time period analog to the Cox model--as well as two additional commonly used models, a logistic and a linear probability model. These statistical models are fit to a relatively large sample (N=2468) of white women and to a relatively small sample (N= 223) of Hispanic women. An identical set of covariates is used for the comparison. Using several tests of goodness-of-fit, all five models capture the general age pattern of early entry into marriage reasonably well, with the proportional rate models closest to the Kaplan-Meier estimates for the whole sample. Regarding the estimates of the effects of covariates, all five models yield quite similar estimates when evaluated at sample means, but of course the linear probability model's estimate deviates substantially from the others at levels far from the means. Although the data demands and complexity of estimation is greater with the proportional rate models, they appear to be the preferred model in terms of their fit with the data. In our comparisons between the two static models, the linear probability model is substantially inferior to the logistic model.
Bibliography Citation
Tuma, Nancy Brandon and Robert T. Michael. "A Comparison of Statistical Models for Life Course Analysis with an Application to First Marriage" In: Current Perspectives on Aging and the Life Cycle, Volume 2, Family Relations in Life Course Perspective. Z. Blau, ed. Greenwich, CT: JAI Press, 1986: pp. 107-146