ABSTRACT
This chapter examines some ways to evaluate reliability and objectivity. The most rigorous way to evaluate the reliability of a measurement is to perform the measurement one time and then administer it again to the same group. Estimates of test-retest reliability assume that there are no major changes in the construct being measured between the two measurements. An alternative method of evaluating reliability is to estimate the consistency of results across items within a test. Reliability is best evaluated with test-retest comparisons, but test-retest measurements are not always practical. The amount of time elapsed between repeat measures is critical. Generally, the shorter the delay between tests, the better the reliability; the longer the time elapsed, the lower the reliability. Test-retest stability gives strong evidence for reliability, but it is time consuming and laborious.
