In this exercise, run a reliability analysis for each of the two factors from the previous exercise. In the first reliability analysis, use all 16 of the negatively worded items that had loadings greater than .30 on Factor I in Exercise B4. Then, do a second analysis for the four items that had loadings greater than .30 on Factor II (you can use either the original or reversed items, it will not make a difference because all items are scored in the same direction). Use the SPSS instructions from Exercise B2 to guide you in doing this analysis. Then answer the following questions: (a) For the 16 items on the first factor, what is the value of Cronbach’s alpha? How does this compare to the value from Exercise B1? (b) Could alpha be increased by eliminating any of the 16 items? If yes, which one? (c) For the four items on the second factor, what is the value of Cronbach’s alpha? Should a subscale such as this one be used to measure Positive Affect? (d) Could alpha be increased by eliminating any of the four items? If yes, which one? (e) What do these analyses suggest about researchers’ use of the full CES-D scale with this population?
