Using SPSS to Understand Research and Data Analysis.

• 5.1c So what are these new variables?

If you look back at Table 4.5 and also scroll through the ezdata.sav file in the SPSS Data Editor, you will see that the original variables shown in Table 4.5 do exist in the new ezdata file, they are just in different columns than you might have expected. Gender does appear in the first column, but the masc1-masc5 variables (the self-ratings on the 5 masculine personality traits) do not begin in the second column. Instead, scroll right and you will see them in columns 18-22. The fem1-fem5 scores appear in columns 23-27.

The variables named masc and fem that you see in columns 2 and 3 are actually new variables that have been created by transformations of the original masc1-masc5 and fem1-fem5 variables. Click the Variable View tab at the bottom of the Data Editor for a better view of all of the variables. Recall that this view lists variables in rows instead of columns (Figure 5.2).

To create the new variables, masc and fem, we used the SPSS Transform procedure to add up the scores on the masc1-masc5 and fem1-fem5 variables to create two variables named masctot and femtot. If you scroll all the way down, you will see these masctot and femtot variables (SPSS tacks newly-created variables at the end of the original data file).

Next, we used the SPSS Transform procedure again to convert the continuous masctot and femtot variables into categorical variables, which we named masc and fem. Last, we moved these two new variables to rows 2 and 3 for easy access later. If these variable transformations sound confusing, don't worry - we will explain all of this in the next section!

You will also see the original soc1-soc3 and task1-task3 variables (employee's scores on social and task leadership skills) in rows 12-14 and 15-17, respectively. We performed similar transformations on the soc1 and task1 variables to create new categorical variables that we named soc and task, and then placed them in rows 4 and 5 for easy access.

Next, you will see the original ach1-ach5, aff1-aff5, and dom1-dom5 variables (employee work motive scores) in rows 28-42. We used the the Transform procedure on these variables to add them up and take their average to create the new variables, nach, naff, and ndom (short for achievement needs, affiliation needs and dominance needs), and we placed these three new variables in rows 6-8 for quick access.

Figure 5.3 graphically depicts these variable transformations. Note that the original variables (in blue) still exist in the data file, and their transformed versions have been given new names (in red), and now exist as separate new variables in the file.

Figure 5.3

You will find the original perform1-perform3 variables in rows 9-11. We have not done any transformations on these variables. Last, if you scroll all the way down, you will see yet two more new variables, sextype and leaderstyle. We also used the Transform procedure to create these new categorical variables, but we will not discuss how we did this in the present chapter. We explain this procedure in a later chapter since it is a bit more complex than the others we describe in this chapter.