HALO BIAS:
Errors which occur when judgements made on an individual’s performance are influenced by the Raters overall impression of the person being judged, rather than the specific characteristics under study1,2. For example, a patient who is recovering from surgery may perform well on some physical tests, and thus score high overall with respect to their recovery, and not required to perform any additional tests, even though they may have performed poorly on those additional tests.
Halo Bias is often referred to as a ‘Halo Effect’ or ‘Illusory Halo’1; as it is suggested that, to some degree, when one performs well in a specific task they will likely perform well in other similar tasks; thus Halo Bias somewhat reflects reality1. Defining Halo Bias as Bias implies this generalization of performance is a systematic error, and thus an unfavourable characteristic to be mitigated, when in fact, it may not be (see Bias (Definition)). Halo Effect may become a Bias, however, when ratings for a specific group of people (e.g. those taking a drug in an unblinded experimental study) are scored differently than other individuals in the same study; and therefore systematic differences in ratings between groups are evident.
Hallo Bias/Effect may occur because Raters are often unable to evaluate individuals beyond more than a few dimensions1. Halo Bias/Effect may be present when the influence of the observer’s recollection or knowledge of findings on a previous occasion affect the ratings of a participant in a study or clinical scenario; in this case Halo Bias/Effect, may be similar to Expectation Bias. Also see: Authority Bias, Therapeutic Personality Bias, Expectation Bias, Information Bias.
References:
1. Streiner DL, Norman GR, Cairney J. Health Measurement Scales a Practical Guild to Their Development and Use. Fifth ed. Oxford University Press 2014. (Link to Reference)
2. Porta M, ed. A Dictionary of Epidemiology. Sixth ed. New York, NY: Oxford University Press 2014. (Link to Reference)