Self-Enhancement Bias and Criteria Specificity
This study varied how many criteria students could use when rating their abilities relative to peers. It examined whether vague traits leave more room for flattering self-assessments than traits evaluated against specified criteria.
THE IDEA
Method
One hundred and fifty-two Cornell University undergraduates provided a percentile ranking of their abilities in several domains vis-à-vis Cornell students. Four specific domains concerned positive traits (talented, athletic, cultured/sophisticated, active in extracurriculars). Three additional questions asked subjects to rank their standing for undesirable characteristics (trouble handling money, disorganized in schoolwork, social anxiety).
Four different versions of the questionnaire were constructed in order to manipulate the number of criteria that subjects could consider in their evaluations. In the unspecified-criteria condition (n = 39), subjects were given no specific criteria to consider. In the six-criteria condition (n = 35), six criteria of evaluation were listed for each trait. For example, in the domain of talent, subjects were presented with this list of potential criteria: (1) paints; (2) good storyteller, comedian; (3) writes stories, poetry, plays; (4) acts, does theater; (5) sings; and (6) does a craft (e.g., pottery, woodwork). Subjects were told that “For each trait or characteristic below, we have listed the specific criteria we want you to consider when evaluating yourself. Use only these criteria when estimating your percentile score.” Subjects in the four-criteria condition (n = 39) were given four of the criteria to consider in their evaluations. Subjects in the two-criteria condition (n = 39) were only given a menu of two. For these last two conditions, three different versions of the questionnaire were created so that each specific criterion (of the six total shown to subjects) was seen by an equal number of subjects. Thus, no specific criterion was ever over- or underrepresented.
When to use it
Education: provide a rubric so students and teachers can compare performance using the same criteria.
Performance reviews: use job-relevant criteria and concrete examples to examine the basis of self-evaluations.
Personal Development: When setting goals, break them down into concrete criteria to avoid overly optimistic self-assessments.
User Research: When asking survey respondents to rate their own expertise, provide specific dimensions to get more reliable data.
FROM THEORY TO EVERYDAY LIFE
In practice
Classroom Performance
A student places themselves in the top 20% for writing without defining what counts as good writing. A rubric covering grammar, structure, and creativity gives them specific dimensions to assess. Whether the new ranking is more accurate requires a comparison with performance.
Driving Skill
A driver who considers themselves above average could assess speed-limit compliance, turn-signal use, and blind-spot checks separately. This makes the basis for the self-rating clearer; the method described here does not report a driving study.
A MOMENT TO REFLECT
Put it to work.
Think of a skill or trait you consider yourself above average in. What specific criteria do you use to make that judgment? How would your self-assessment change if you narrowed those criteria?
Consider a domain where you feel average or below average. Could it be that you are using different or fewer criteria than others?
How might providing specific evaluation criteria in a workplace or educational setting affect self-assessments of performance?
Often confused with
CONTINUE THE CONVERSATION
Source texts
- Heuristics and Biases The Psychology of Intuitive Judgment (Thomas Gilovich, Dale Griffin, Daniel Kahneman)