B2. In this exercise, you will create a new variable (call it hprobgrp) that divides the sample into three groups based on number of housing problems. The new variable will be coded 1 for no housing problems, 2 for one housing problem, and 3 for two or more housing problems. Click Transform (upper tool bar) ➜ Recode ➜ Into Different Variables. This opens a dialog box where you can enter the variable housprob into the field for Numeric Variable.

Then, on the right, type in the name of the new variable hprobgrp, and then click the pushbutton Change, which will insert the new variable into the first field. Now click on Old and New Values, which brings up a new dialog box.

In the field for Old Value, enter 0 (zero problems), then on the right under New Value, type in 1 (for code 1 in the new variable), and click Add. Repeat this process, this time using the code of 1 for housprob and 2 for hprobgrp, and the click Add again. Finally, on the left click Range (for range of values), and enter 2 and 8—all codes between 2 and 8 for the variable housprob. Then enter 3 for New Variable, and click Add and then Continue. Then click OK, and the process is complete. The new variable will be added as the last variable in the file. Add labels for the codes to make reading output easier by clicking on the Variable View tab at the bottom of the screen. Then run Frequencies for hprobgrp, and determine how many women are in each of your three housing problems groups.

Check to make sure you have correctly created the new variable by comparing the output with output from Exercise B1. Comment on how the distribution of values could affect decisions relating to the homogeneity of variance assumption for ANOVA.