(b)
Would a dependent or independent t test be appropriate?
Why? I would use dependent
because we are comparing the same group, but dividing by employment status.

10
Now run a t test to test the hypothesis, using the dataset “
Statistics Polit SetC data.sav
”:
Analyze→ Compare Means→ Independent Samples T Test.
Move the variable
cesd
(variable #35) into the slot for Test Variable(s).
Then move the variable
worknow
(variable
#6) into the slot for Grouping Variable.
To run this analysis, you must know how the group
variable is coded.
For
worknow
, a code of 0 indicates that the woman was not working, and
a code of 1 was used to code those who were employed.
Click the “Define Groups”
pushbutton and on the next dialog box enter 0 for Group 1 and 1 for Group 2.
Then click
Continue.
Click the “Options” pushbutton and make sure that the computer will compute a
95% CI.
Then click Continue and OK to run the analysis, and answer these questions:
a.
How many women were employed versus not employed in this sample?

c.
What is the value of the F statistic for Levene’s test, and what is the level of significance?

d.
Which formula for t is appropriate in this situation – the pooled variance (equal variance
assumed) or separate variance formula (equal variance not assumed)?
Why? Because the
sample sizes are not equal, the separate variance formula should be used.
e.
What is the value of t for the appropriate test?
6.954 What is the level of significance?

4

f.
What is your conclusion? Is your research hypothesis supported? Write a paragraph
reporting the result. For t at 95% CI, the chart states it was 1.96, while the computed t
was 6.954.
Since the T value from the table is lower than the calculated, we can reject
the null hypothesis.
The data concludes the alternative hypothesis can be accepted.
Employment has an effect on depression levels for this group.
g.
Attach the SPSS printout.

5