- (E) In correlating the two sets of results, Jupiter is measuring the test’s reliability. This type of
reliability is known as test-retest reliability. Since perfectionism should not change over the course
of the month, a low correlation would indicate that the test was not reliable. Outliers are extreme
scores, and Jupiter does not appear to be looking for them. To standardize the test, Jupiter would
have to give it to a standardization sample and then look to assemble a group of questions the
yielded a normal distribution of scores. Nothing in the question suggests Jupiter believes the mean
level of perfectionism should have changed, and the correlation will not necessarily show whether
it has. Finally, even though Jupiter’s work will show whether or not the test is reliable, it will not
show whether or not it is valid. Validity is a measure of accuracy—in this case, whether the test
actually measures perfectionism.
marvins-underground-k-12
(Marvins-Underground-K-12)
#1