In: Statistics and Probability
Assuming that that the two population are normally distributed,
Let denote the variances within groups / errors of the subjects under the Treatment and Placebo groups respectively.
Given: ,
We have to test:
Vs
The appropriate statistical test to test the above hypothesis, would be an F test for equality of variances; for which the test statistic is given by,
with critical region given by
Substituting the given values,
Comparing the test statistic with the critical value F for at 5% significance level which is equal to:
F(0.05,20,20) = 2.124 (From F table)
Since, F = 7.61 > 2.124 lie in the critical / rejection region, we may reject H0 at 5% level. We may conclude that the data provides sufficient evidence to support the claim that the treatment group has errors that are vary significally more than the errors of the placebo group.