Pass or fail? Binomialrelated hypothesis testing and confidence intervals using paired samples - Improving the User Experience through Practical Data Analytics - page 131

We now arrive at the “Tests for Several Related Samples” dialog box, as shown
in Figure 5.10 . Check off the small box to the left of “Cochran's Q” (see horizontal
arrow in Figure 5.10 ), and uncheck the small box to the left of “Friedman,” which is
the default (see vertical arrow in Figure 5.10 ).
Now we drag over the two “variables” to the “Test Variables” section, as shown
in Figure 5.11 .
Click on “OK” to obtain our output, as shown in Figure 5.12 .
Notice that, of course, we obtain the same value for Cochran's Q (which equals
0.333; see top arrow in Figure 5.12 ) and the same p -value, which equals 0.564 (see
bottom arrow in Figure 5.12 ) as we did in Excel. As a consequence, we reach the
same conclusion of accepting H0, hence concluding that the completion rates cannot
be said to differ for the two tasks.
FIGURE 5.10
“Test for several related samples” dialog box; SPSS.