Reliability refers to the consistency of a test in measuring the intended construct. A test-retest method is used to determine reliability by administering the same test twice to students and computing a rank correlation coefficient, such as the Spearman Rho, based on comparing the rank orders of student scores between the two test administrations. The example shows student scores from an initial (FA) and subsequent (SA) administration of a test, with their ranks and the differences between ranks used to calculate the standard deviation.