What was the range of missing data for individual items?.
1. Comment on the researchers’ decisions in the research example at the end of this chapter (Cˇrncˇ ec et al., 2008). What, if anything, would you recommend doing differently?
2. For these exercises, you will be using the SPSS dataset Polit2SetC. This file contains responses to individual items on the Center for Epidemiologic Studies— Depression (CES-D) Scale. This scale presents respondents with 20 statements about their mood and feelings. They are asked to rate each item for its frequency in the prior week on the following scale: 1 (rarely or never—less than 1 day), 2 (some or a little—1–2 days), 3 (occasionally or a moderate amount—3–4 days), or 4 (most or all—5–7 days). The CES-D has been used in thousands of studies, and has undergone rigorous psychometric testing. Still, it is worthwhile to assess its dimensionality and psychometric adequacy for a population of low income minority women. Let us begin by running basic frequency information. In the file, you will find that there are 24 CESD items because four items (item 4, 8, 12, and 16) are worded positively (e.g., “I was happy,”), and so to consistently measure depression these items have to be reverse scored. We have done this for you. Run basic descriptives for the 24 variables (starting at cesd1 and ending at cesd20) and then answer the following questions: (a) Were there any items that did not have a full range of responses, from 1 to 4? Were there any outliers? (b) What was the range of missing data for individual items? (c) Comment on the similarity or differences in means and SDs across items. (d) Examine the frequency distribution information for item 4 and item 4 reversed. Does it appear that the reversal was done properly? (Note: If you will be doing the exercises on Missing Values, keep the output from this exercise for later reference.)