Refer to the Week 5 t Test Exercises and follow the directions to perform a t test.

Part I: Independent Samples t-test

The hypothesis being tested is: Women who are working will have a lower level of depression as compared to women who are not working.

1. In the Polit2SetC dataset, 773 women were employed and 299 were not employed.
2. The total sample size is 1072.
3. The mean (SD) CES-D scores for the employed group is 16.18 (11.53) and for the non-employed group is 19.80 (11.63).
4. The Levene’s statistic is 11.161, and the p-value is .001, indicating that the assumption of homogeneity of variance is violated. Equal variances are not assumed.
5. The t-statistic is -4.222 with 1070 degrees of freedom and the p-value is < .001.
6. The data supports the hypothesis that women who are working have a lower level of depression than women who are not working.

Part II: Paired Samples t-test

The hypothesis being tested is: Women who reported depression scores in wave 1 and wave 2 of the study did not have a significant difference in their level of depression.

1. The total sample size is 1144.
2. The mean (SD) CES-D scores at wave 1 is 16.12 (11.55) and at wave 2 is 14.51 (11.21).
3. The mean difference between the two time periods is 1.61.
4. The t-statistic is 3.999 with 1143 degrees of freedom and the p-value is < .001.
5. The data does not support the hypothesis that there is no significant difference in depression scores between wave 1 and wave 2.

Part III: Independent Samples t-test for three outcomes

The outcomes variables are CES-D Score (cesd), SF12: Physical Health Component Score, standardized (sf12phys) and SF12: Mental Health Component Score, standardized (sf12ment).

The hypothesis being tested is: Women with no high school credential will have worse CES-D, physical health, and mental health scores than women with a diploma or GED.