In this article, the last in a series of four on performance appraisals and how they relate to safety, I want to discuss six psychological factors that prejudice typical performance evaluations. Remember, generic evaluations and rankings won't improve competence - in safety or any other area. Still, becoming aware of these factors can make your evaluations more accurate.

Leniency errors

There's a tendency to evaluate everyone favorably. Why? Because managers want to avoid negative confrontations. They anticipate employees protesting low evaluations as unfair and biased. And unless the evaluation is free from subjective interpretation and based on periodic objective observations of the employee's behaviors, a strong case can usually be made for an unjust and corrupt evaluation.

But if everyone gets a similar above-average score on a generic performance evaluation, the rating process has limited value. This is a prime reason for eliminating the traditional evaluation and ranking component of a performance appraisal system.

Halo effects

These occur when an initial positive impression of an individual leads to perceiving everything the person does in a favorable light. The opposite bias is termed a "devil effect," and occurs after we form a negative overall impression of a person. Then we're more attentive to negative than positive aspects of everything the person does.

This bias occurs because initial impressions usually have the most powerful impact and influence all subsequent observations. In other words, our perceptions of people are often biased by an attempt to confirm our first opinions.

There are two lessons here. First, since first impressions are so influential, work hard to generate a positive one with coworkers. And try to observe others' performance as if you're seeing them for the first time. Recognize that prior assessments predispose the way you see a person today.

Affective reactions

You've likely experienced this prejudice many times. Perhaps you've tried explicitly to separate your personal feelings for an individual from an evaluation of that person's performance. But a large amount of psychological research indicates that this is extremely difficult, even when evaluators are aware of this bias.

Consider that your observations of another person's performance are influenced to some degree by your affective or emotional reaction to this person. Consider, too, that your formal evaluation of this person occurs days, weeks or months after your observations, allowing plenty of time for emotional feelings to distort your memory of what you saw, as well as your interpretation of the relevance, utility, and validity of what you remembered seeing.

Attributional bias

Suppose you need to evaluate two people, and in your judgment both achieved the same performance output. But you believe one of these individuals loafed along at half speed, coasting to the observed performance level. In contrast, the other person is less talented and reached this level of output by working much harder. Would you give both of these individuals the same performance rating?

Research indicates that most people would assign a higher rating to the second individual - the one who put out the most effort and exceeded beyond expectations. But is this fair? Should effort count? If the evaluation is based on external performance output only, subjective judgments you attribute to internal motivations are irrelevant and should not influence the assessment.

Empathy errors

Being human, we appreciate and like people who are similar to us. We can readily put ourselves in these people's shoes, and understand (we think) what motivates them. So it's not uncommon to assign higher ratings to people who are similar to the evaluator in various ways, compared to people who are dissimilar.

This bias links to the prior two factors - affective and attributional prejudice. We are apt to like people who are similar to us. And when these people perform well, we are likely to attribute the behavior to positive internal motivation. After all, these people are similar to us, and we see ourselves as having more self-discipline and internal drive than most others in our work setting, right?


I wrote about stereotyping in my September 1999 contribution toISHN, when I discussed the problem of mindless labeling of people and introduced the related concept of "premature cognitive commitment." In that article, I proposed that the key to reducing prejudice is to make more - not fewer - distinctions between people. Becoming more mindful of the numerous differences among individuals and how these differences fluctuate according to time, place, and social context, makes it difficult to attach labels to people.

Yet, let's face it, we do put people in categories. Such discrimination is even facilitated by popular personality tests. This only contributes to this stereotyping bias. Becoming more mindful of the vast differences between individuals helps to reduce the mindless categorization of people.

No easy task

Becoming aware of all six of these psychological factors that bias our evaluations of others can reduce, but not eliminate, their influence. Really, we should try to eliminate the type of generic evaluation and ranking that occurs in the standard end-of-the-year performance appraisal.

An effective performance appraisal customizes behavioral objectives and feedback per each employee, and avoids these prejudicial factors. It's not easy to conduct an effective performance appraisal. There's no quick fix; no middle ground. But when it comes to evaluating human performance in an attempt to improve competence in safety and other areas, don't do it if you can't do it right.