In our study of comparisons between the means of two groups, it was noted that the most common form of the two-group t-test for the difference between two means assumes that the population variances for the two groups are the same. One experimenter, using a control condition and an experimental condition in his study of drug reaction, wished to verify that this assumption held, that is, that the treatment administered affected only the mean, not the variance of the variable under study. From his data, he calculated the variance of the experimental group to be 25.8 and that of the control group to be 20.6. The experimental group had 25 subjects, and the control group had 31. Can he proceed to use the t-test, which assumes equal variances for the two groups? Use α = 0.10.

Contact us for the solution