Requested White Paper
Retesting: What is the Right Time Period?
Administering a test costs time and money, so giving a test multiple times when it will not significantly change the outcome for a candidate is a waste of both time and money. Moreover, there's always the possibility that practice effects could become a factor if the test is taken again within a short period of time, and therefore raise the score of the individual without any “true” change in the attributes that you are testing for in the first place.
A basic property of a good test, or of the scores generated by a good test, is adequate levels of reliability or consistency. Test-retest reliability is the measure of a test’s temporal consistency when the test is taken by the same individual (or group) at least twice. Traditionally, test-retest reliability is demonstrated by correlating the individual scores from Time 1 with the individual scores from Time 2. Test reliability is never perfect, there are always things that affect a person’s score from one time to another.
