Halo Effect

A halo effect is a systematic error influencing an evaluation outcome that is attributable to an interaction between the evaluator and the object of evaluation. In the simplest context, an outcome from an evaluation can be decomposed to reflect the influence of at least two separate components and the interaction of these two components. One component is the evaluator who is responsible for rendering a perception. The other component is the phenomenon being evaluated and [Page 445]could include a person performing certain job assignments, as involved with performance appraisals, or a focal position within an organizational hierarchy, as involved with a job analysis, to mention two possibilities as the objects of evaluation.

Outcomes from the evaluation process yield an observed perception about the employee or the position being evaluated. Because this outcome is based on perceptions of the evaluator, it is subject to several different types of perceptual errors. Some of the potential perceptual errors in outcome measures may be attributable solely to the evaluator, while some of the other potential perceptual errors in outcome measures may be attributable to an interaction phenomenon involving an evaluator and an object of evaluation (halo effect).

With respect to those perceptual errors associated solely with the evaluator, several common types exist. For example, some evaluators may evaluate all employees artificially high within the appraisal process (leniency error) or different jobs of varying value equally within the job analysis process (central tendency error). That is, these types of perceptual errors within the evaluation process are constant across all objects within the evaluations of a particular evaluator.

In contrast to the above-illustrated perceptual errors attributable solely to the evaluator is the halo effect. The halo effect involves an interaction between the evaluator and a specific object of evaluation. As such, errors in the perceptions of an evaluator are not systematic across all objects being evaluated.

To illustrate, only certain employees are evaluated artificially high, or only specific jobs are erroneously overvalued with a halo error. Within the context of a halo effect, the evaluator's perceptions are tainted by specific attributes associated with persons or programs that are unrelated to the purpose of the evaluation process. If an evaluator appreciates a specific trait of a particular employee such as appearance, then this appreciation influences the ratings of this particular employee across all aspects of the job performance evaluation on dimensions unrelated to appearance. Likewise, a job analysis can be unduly influenced by a specific job component associated with a particular job (e.g., public exposure), and this influence can color the analyst's evaluation of all job components comprising a job.

To identify a potential halo effect associated with an evaluator by object interaction, descriptive statistics are calculated for each object of evaluation nested within every evaluator. Classical indicators of a halo effect are a high mean rating across all dimensions of a particular individual/job being evaluated accompanied by small standard deviations. This particular figuration of evidence is far from conclusive in proving a halo effect but does signal for further investigation on a case-by-case basis.

...