B5. For the next analysis, run a two-way ANOVA in which the dependent variable will again be satovrl, the women’s overall degree of satisfaction with their material well-being. The two dichotomous independent variables will be whether or not the woman was working at the time of the interview

(worknow, Variable #8) and whether or not she was receiving cash welfare assistance (cashwelf, Variable #19), both coded 1 for “yes” and 0 for “no.” Run two-way ANOVA through Analyze ➜ General Linear Model ➜ Univariate. In the opening dialog box, move satovrl into the Dependent List, and move worknow and cashwelf into the Fixed Factors list. Click the pushbutton for Model and unclick “Include intercept in model” at the bottom of the next dialog box. Then click Continue. Next, click the pushbutton Plots on the initial dialog box. Move worknow into the Horizontal Axis slot, and then move cashwelf into the Separate Lines slot. Click the Add pushbutton, then Continue. Next, click the Options pushbutton on the original dialog box and select the following options: Descriptives, Estimates of Effect Size, and Homogeneity Tests. Click Continue, then OK, and answer the following questions:

(a) What are the null hypotheses being tested?

(b) Which of the four groups created by the 2  2 design was, on average, least satisfied with their housing? How many women were in that group?

(c) Which group was most satisfied? How many women were in that group?

(d) Are the variances homogeneous?

(e) In the ANOVA, which (if either) main effect was statistically significant?

What were the probability levels?

(f) What were the values of eta-squared for the main effects? How would you describe these effect sizes? (g) Was the interaction effect significant?

What was the probability level for the interaction? Describe interaction effects as displayed on the Profile Plot.