The data in Table 10.9 consist of general severity rating scores from 24 participants after each participant receives therapy from one of 6 therapist trainees. Three of the trainees are males while the other three trainees are females. In this sense, the trainees are nested within their particular gender.
The hypothesis of primary interest is determining the effect of gender on severity score ratings. As of this writing, there is no procedure using the SPSS point and click method that provides the correct statistical tests for the gender effect. One can utilize the Linear Mixed Model procedure in SPSS to obtain correct results for the test of the trainee within gender effect, but cannot obtain the correct test for the gender effect through point and click. Thus, the appropriate method used to perform an ANOVA on nested data in SPSS is to use the syntax. See the SPSS syntax to obtain the appropriate statistical tests for Table 10.9.