| Abstract: | The 1979 National Longitudinal Survey of Youth (NLSY) provided data for the analysis of the impact of self-esteem, locus of control, attitudes towards women's family roles, work, and school on the probability of a nonmarital birth. The study avoided methodological problems of prior studies by using a national data base with attitude measured before nonmarital childbearing occurs. No prior studies included these 5 factors. Previous research on attitudes and nonmarital childbearing is summarized. Theoretical models imply that self-esteem, high educational goals, and an internal locus of control are associated with a lower likelihood of a nonmarital birth. The conceptual model for this study is described. It does not include how attitudes develop or estimate the linkage between family background variables and attitudes. A reduced from approach was used and controls were included for 6 family background and personal variables. The sample of 1184 girls was restricted to those aged 14 or 15 years in 1979 who were never married or had a child. 16.9% of the sample had a nonmarital child by 19 years. Measurement of explanatory variables is indicated as the Rosenberg Self Esteem Scale and the Rotter scale for internal-external locus of control. A 7-item scale measured women's roles, an 8-item scale measured attitudes toward school, and 3 items assessed views on the importance of improving one's employment prospects. Background control variables were race/ethnicity, mother's education, presence of welfare income, family income, family structure, and religiosity. The standard logistics technique was used to estimate the logarithm of the odds of having a nonmarital birth as a linear function of both attitude and family background variables. Models were estimated both with and without the educational expectation variable. The results appear to indicate that self-esteem and attitudes toward school are associated with nonmarital childbearing, as predicted by theory. Locus of control does not systematically affect the likelihood of nonmarital childbearing, nor do family and gender roles and attitudes toward work. Educational expectations are not associated, but may act as an intervening variable. |
