A question that often comes up when implementing an assessment program is whether job candidates should be allowed to re-test. In most situations, it is good policy to permit it. For example, an individual may not have been feeling well, was excessively anxious, or for some other reason wasn’t able to show the best of his or her abilities on the day of the first test. Additionally, it is often important to allow re-testing for the sake of an organization's reputation because prohibiting it will often cast it in a bad light. Indeed, re-testing is also encouraged by professional standards as long as it is done in a reasonable way that protects the integrity of the assessment program.

That being established, the question then becomes what impact does this have upon test scores? Deidra Schliecher and associates recently published a paper in the Journal of Applied Psychology that analyzed testing data from over 2000 candidates who applied for a professional-level government service position in the US. Looking across the battery of tests that were administered, scores increased by .1 to .4 of a standard deviation depending upon the test. (This is consistent with other research indicating that scores do tend to increase somewhat upon re-testing.) Complicating matters a bit, there were some sub-group differences in the rate of improvement. For example, whites improved more than African-American and Hispanic candidates (though African-Americans improved more than other groups on the structured interview.) Also, younger candidates improved more than older candidates while women improved more than men.

The end result is that re-testing (while considered a reasonable policy) can slightly increase the level of adverse impact for some groups though this will vary as a function of factors such as the selected cut score and differential levels of re-testing between groups. Of course, this study is limited by the fact that this is only for one position, and these sorts of findings can vary significantly based upon the specific population at hand.  However, the takeaway message is that all assessment programs need to be monitored on an ongoing basis, not simply for compliance purposes but to ensure they continue to meet the needs of the organization.

Reid Klion