Comparing more than two means: one factor ANOVA with a withinsubject design - Improving the User Experience through Practical Data Analytics - page 172

FIGURE 7.3
Univariate dialog box for one-factor ANOVA; SPSS.
The “Respondent” variable goes to the random factor box (horizontal arrow in
Figure 7.4 .)
We are now ready to click on “OK.”
The output is presented in Figure 7.5 .
The key quantity is the p -value for “Task,” which is 0.000 in Figure 7.5 (see
arrow). Recall that SPSS rounds all p -values to three digits.
So, the p -value for “task” is way below the 0.05 cutoff point, and we conclude
that the mean ease-of-use rating is not the same for the four tasks.
We might note that the p -value for “people” is 0.081 (see oval in Figure 7.5 ). This is
not below the 0.05 traditional cutoff point, indicating that we conclude that the mean for
each respondent , averaged across all four tasks, cannot be said to be different (although
it is close—the p -value is not much above the 0.05 cutoff). This latter point is not of
prime interest to us, but it does indicate that we do not have enough evidence to refute
that the people, on average, are consistent in their evaluations of the ease-of-use of the
four tasks. Of course, this does not indicate directly that each person rates every task the