A5. Assume that a researcher has conducted a pilot intervention study and wants to use the pilot results to estimate the when there are more than two groups or observations, the Friedman test is...

A5. Assume that a researcher has conducted a pilot intervention study and wants to use the pilot results to estimate the when there are more than two groups or observations, the Friedman test is appropriate. These tests are the nonparametric analogs of the paired t test and repeated measures ANOVA, respectively.

number of participants needed in a full-scale study to achieve a power of .80 with a = .05. For each of the hypothetical pilot results presented below, how many study participants in each group would be needed in the larger study? (Note that in using Table 3, it makes no difference which of two groups is considered Group 1 or Group 2.)

(a) Experimental group: 45%, Control group: 65%

(b) Experimental group: 15%, Control group: 5%

(c) Experimental group: 20%, Control group: 35%

(d) Experimental group: 60%, Control group: 75%