The purposes of this study was to identify the significant predictors of depression. The data were collected using questionnaire from the sample of 114 middle-aged women and 125 middle-aged men. The data were collected through self-report questionnaires, which were constructed to include stress, self-esteem, and depression. Data were analyzed using frequencies, means, $X^2$-test, Analysis of Covariance, Pearson's correlation coefficients, and multiple regression analysis with SPSS 18.0. The significant predictors of depression for middle-aged women were stress(${\beta}=.387$, p<.001), self-esteem(${\beta}=-.249$, p<.05), health perception(${\beta}=-.191$, p<.05), explaining 43.1% of the variance in depression. The significant predictors of depression for middle-aged men were self-esteem(${\beta}=-.429$, p<.001), stress(${\beta}=.322$, p<.001), explaining 56.0% of the variance in depression.