Results

In the sample of graduate students approximately 21% indicated they did not have health insurance. Of the students who did not have health insurance the most frequently cited reason was financial constraints (64%) followed by aging out of parents’ coverage (19%).

Descriptive statistics and correlations between the variables were calculated (Table 1). A series of t-tests revealed no significant gender differences in any of the outcome variables. The data was normally distributed with skewness and kurtosis values within the +/- 1 range. The result of a Box’s test for equality of covariances indicated equivalent variances across outcome variables, F(6, 4734) = .17, p = .99. The results of a MANOVA Wilk’s lambda test indicated that there were significant differences between the groups on depression, stress, and anxiety, F(3, 61) = .334.67, p < .001. Discriminant analysis suggested that depression was most strongly associated with differences between groups (.67) followed by stress (.25) and anxiety (.01). While significant differences between groups on the individual variables cannot be determined using a MANOVA, individuals without health insurance had scores in the expected direction with greater depression, stress, and anxiety means.