In 1920, psychologist Edward Thorndike published an article with the telling title "A Constant Error in Psychological Ratings." He studied how commanders rated their soldiers on various qualities: physical fitness, intelligence, leadership, and character. It would be logical to expect that these ratings would be fairly independent of each other—a person could be physically strong and average in intelligence. But the ratings turned out to be linked far too closely. A soldier whom the commander considered good received high marks on almost all counts. The one he disliked received low ones. Thorndike called this the halo effect.
Half a century later, Richard Nisbett and Timothy Wilson demonstrated just how deeply this mechanism operates. Students were shown a videotaped interview with a lecturer who spoke with an accent. One group saw a version where he behaved warmly and amiably, while the other saw one where he was cold and aloof. They were then asked to evaluate his appearance, mannerisms, and accent—that is, things that were identical in both recordings. The first group found them pleasant, while the second found them irritating. And when the students were asked whether their overall attitude toward the lecturer had influenced these evaluations, they confidently replied that it had not.
This is the most troubling aspect of the effect. It is not felt from the inside. A person is sincerely convinced that they are evaluating each quality separately, whereas in reality, they are projecting an overall impression onto everything at once.
In HR, this mechanism is present almost everywhere one person evaluates another across multiple criteria.
Annual performance appraisal. If an employee performed well in one high-profile project, their ratings for communication, initiative, and meeting deadlines are pulled upward as well—even though there might have been real issues on those counts. The reverse version works too, sometimes called the horns effect: a single noticeable mistake pulls everything down.
Job interview. A confident demeanor, a strong resume, or a recognizable name of a past company creates an impression through which all of the candidate's answers are subsequently interpreted.
Promotion. A person perceived by leadership as strong is given the benefit of the doubt regarding skills that no one has actually tested.
The problem is not that evaluators are careless. The problem is that the evaluation format is usually structured in a way that triggers a halo: one person fills out all the criteria consecutively for a single employee, and the first impression manages to set the tone for all the other points.
What helps. Evaluating by criteria rather than by people: first evaluating one criterion across all employees, then the next, rather than completing the entire form for one person at once. Requiring a concrete behavioral example for each rating—without an example, the rating is not accepted. Separating the evaluation of different qualities among different evaluators whenever possible. And conducting calibration sessions where suspiciously uniform profiles are examined separately: if a person's scores are all uniformly high or uniformly low, this is a reason to check whether the same single impression was being measured over and over.
The more detailed the questionnaire is, the more convincing it looks. But if it is filled out by a halo, twenty criteria turn into twenty ways of repeating a single opinion.
#HR #ОценкаПерсонала #Психология #Грейдинг
📱 LinkedIn (https://www.linkedin.com/company/gradex-uz/) | 📱 Instagram (https://www.instagram.com/gradex.uz/) | 📱 Telegram (https://t.me/+KKX73T0bSyAyNjQ6)