Data Analysis
The analysis were conducted using SPSS and STATA 10.0. A two-tailed p value less than.05 was the criterion for statistical significance. The initial data analysis was descriptive in nature. The stress measures(perceived stress, prenatal distress, and CRH) were summarized at each data collection point for the three groups of Black women. Pearson's product moment correlation coefficient was used to estimate the strength of the linear association between measures. One-way analysis of variance with post hoc Bonferroni-adjusted tests were used to identify mean differences in stress levels between groups at each time point. Differences in stress over time and differences between groups were examined using generalized estimation equations(GEE). These regression models were constructed for each stress variable and included time as a covariate; if the regression coefficient for time differed significantly from zero, then this would indicate that there was a significant change over time for the stress variable under consideration Furthermore, a positive(negative) value for the regression coefficient would correspond with an increase(decrease) in stress. In addition, the GEE models were used to compare change in stress levels between two groups: preterm delivery(PLPD) versus term delivery(PLTD+TLTD). These models included time, an indicator variable for preterm delivery(PLPD), and a preterm delivery by time interaction term. An interaction term that differed