Social policy evaluation: How did the 1996 Welfare Reform Legislation influence unmarried Teen Birth rates in Texas Counties

Implementation of the 1996 Welfare Reform Legislation radically changed a 61-year-old policy of providing welfare benefits to eligible low-income mothers and their children. Despite a lack of evidence, current welfare policies are being implemented based on the belief that welfare restrictions and sanctions can change the reproductive behavior of women, particularly teens. The purpose of this quasi-experimental study was to examine and describe the influence of the 1996 welfare reform legislation on unmarried teen childbirth rates in Texas counties and to examine the effect of the counties, teen population density, ethnic homogeneity, and welfare participation on the rate of change in unmarried teen childbirth rates post-welfare reform.

In this study, the 254 counties in Texas represented the population under examination. The variable of interest was the number of births to unmarried women less than 20 years of age from January 1, 1994 through December 31, 1999. Two-stage growth curve modeling was used to examine the pattern of the unmarried teen birthrates pre- and post-welfare reform. During the first stage of the modeling, trajectories that represented the quarterly unmarried teen birthrates were estimated using multiple regression analysis. A paired-sample t-test was used to test for across county differences in the pre- and post-welfare reform slope estimates. In stage two, hierarchical linear regression was employed using teen population density, ethnic homogeneity, welfare participation, and the post-welfare reform intercepts to determine if they could explain the change in the unmarried teen birth rates post-welfare reform.

The results of the growth curve modeling found that 17 Texas counties had statistically significant changes in their slopes and/or intercept estimates post-welfare reform as compared to the estimates pre-welfare reform. The results of the paired sample t-test revealed no significant difference between the mean of the pre-welfare reform slopes (x¯ = .047) and the mean of post-welfare reform slopes (x¯ = .045), t(253) = .053, p = .958. The results of the hierarchical linear regression analysis indicated that the estimate of the post-welfare reform intercept and welfare participation were the only variables that contributed significantly toward predicting the slope of the unmarried teen birth rates post-welfare reform.