Data Analysis
Percentages and frequencies described the younger
(13–16 years) and older (17–19 years) adolescent
demographically and the prevalence of depressive
symptoms and VE by gravidity and parity. For descriptive
purposes, depressive symptoms were defined
as minor (10–12) and major (13 and above)
per the authors of the EPDS. Differences for repeat
pregnancies among groups that reported depressive
symptoms or VE were examined using analysis of
variance (ANOVA). The outcome variable of repeated
pregnancy (measured as gravida and parity)
was recoded as 0 and 1 because of the small number
of multigravid and multiparous adolescents, with
0 equal to one pregnancy/living child and 1 equal
to two or more pregnancies/living children. Logistic
regression was performed to determine the extent
to which repeat pregnancy, measured as gravidity
and parity, was predicted by depressive symptoms
and VE. The model alphas were set to .10 as is recommended
by Hosmer and Lemeshow (2000) for
pilot data submitted to logistic regression. Multicollinearity
diagnostics, including tolerance and
variance inflation factor values, were well within
normal limits.