Target Variable field,

Now we will look at how many missing values individual participants had for these 20 CES-D items. To do this, we need to create a new variable (we will call it misscesd) that is a count of how many values are missing for each person. Select Transform ➜ Count Occurrences of Values Within Cases. In the Target Variable field, enter misscesd; you can add a descriptive label such as “Number of missing CESD items.” Then move the 20 individual CES-D items into the field labeled Numeric Variables. Click the Define Values pushbutton to get to the next dialog box. In the section labeled “Value,” select System- or user-missing, then click the Add button. Click Continue, then OK. Next, run a frequency distribution and descriptive statistics for the new misscesd variable, and use the output to answer the following questions: (a) How many women answered all 20 items? (b) How many women answered none of the 20 items? (c) What percent of cases had missing values on more than half of the items? (d) What was the mean, median, and modal number of missing items? (e) If we used the standard of imputing missing values only for items with no more than 25% of the items missing, for how many cases would we impute missing values? If we then computed CES-D scales scores, for what percent of the cases would we not be able to compute a CES-D score?