When and if the researcher would test the boys before and after treatment for depression this would skew the two tests and the results would be more positive for the treatment would affect the outcome of the second test. The first test when the boys were depressed they would not put all their thinking skills into doing an honest job for the way they are feeling at that time. When testing boys and girls with depression the researcher must take into consideration other circumstances into performing the test.
The repeated measures test is still the best t-test to use for this test will give the researcher a better look at the emotions of the boys through observation while performing the test. They can also observe facial and body movements of the participants. In all psychological testing researchers must strive for validity and to make sure the criteria is met before performing the test along with the reliability of the results found. The researcher can also use the single sample t-test when checking the boys after treatment they received for depression. They also need to find the degrees of freedom dealing with the participants from the previous sample. The researchers could also perform a paired samples t-test between the girls and the boys to see about levels of self esteem and with the paired samples t-test this is when Cohen’s d is used. The tester will also need to calculate the effect size and get the sample mean difference.
When testers give these types of tests they should make sure that they are appropriate, meaningful and useful. (Gregory, 2004) When dealing with the statistics and the t-test researchers should strive to make ideas relevant to the test that is being performed. When considering which test or tests to be done one find the correct charts to make the necessary calculations.