Validity, Halo Effect

The halo effect refers to the idea that a person or attribute considered highly (or lowly) valued in one aspect becomes highly (or lowly) valued in [Page 1825]some other aspect unrelated to the original assessment. Consider, for example, a person who receives a high (or low) evaluation in some domain, such as the ability to shoot a basketball, based on performance. The “halo” of a positive evaluation as a basketball shooter might then be applied to the ability of the person to drive a motorcycle. The evaluator, without any evidence or performance information regarding motorcycle driving, assigns a high (or low) value to that person based on the person’s basketball performance. The justification for the positive (or negative) evaluation of the ability to operate a motorcycle fails to reflect any value other than the positive (or negative) evaluation in the other domain, basketball ability.

The “halo” refers to the aura of perception that the evaluator selectively interprets information so that all actions of the person become interpreted consistent with the original evaluation. A good person can do no wrong or a bad person can do no good. Halo effects serve to some extent as the basis of stereotypes, employ selective perception, and often can be viewed as related to attribution theory.

As a research endeavor, any design or analysis using a rater to make multiple evaluations of the same person may reflect a common set of assumptions or valence. An evaluator or observer may provide a positive evaluation on one element and that evaluation can impact all subsequent assessments. The use of a single person to perform multiple evaluations risks this kind of problem, that is, having one rater provide both sets of evaluations. The solution involves finding a means to generate independent assessments.

One outcome of the halo effect is a higher reliability among assessments of the rater across situations. The impact of the rating issue, when it occurs, has the “positive” consequence of raising cross-situational consistency. However, the appearance of gain across situations becomes illusionary and an outcome of the measurement if a comparison to independent evaluations demonstrates significant departure from the shared assessment.