The person-situation debate was sparked by an influential book by Walter Mischel in 1968, in which Mischel attacked the trait approach to personality. Mischel held that ratings on traits fail to predict actual behavior. Epstein argued in 1983 that Mischel's analysis was flawed. Although traits may fail to predict single behaviors, traits are good at predicting aggregates of many behaviors. For example, someone who is conscientious will likely study for most exams. However, that person may not study for a particular exam because of an illness. Therefore, traits are not good at predicting single behaviors, but are good at predicting aggregates of behaviors.
Brody and Ehrlichman (1998) devised the following hypothetical study to test whether traits are good at predicting behaviors across not only many examples of the same situation, but also across different situations. This test should look familiar--it is simply the construction of a hierarchical model, like Eysenck's PEN model.
- Obtain measures of behavior for a group of individuals in each of twenty situations that are assumed to be relevant to the trait of conscientiousness.
- Assume further that each individual has been observed several times in a situation and that the measure or index of behavior in a situation for each individual is baseed on an aggregate index of behavior.
- Divide the set of situations into two arbitrary groups of ten each.
- Obtain two aggregated indices of conscientiousness for each individual by averaging the person's score for each of the two groups of ten situations. Note that each of these aggregated indices of conscientiousness is cross-situational, based on aggregating measures of behavior in different situations.
- Obtain a correlation between the two indices. Can you guess the value of this correlation? It would probably be relatively hight--perhaps close to .7. This correlation would imply that individuals who behave in a conscientiousness manner in one set of situations are likely to behave in a conscientiousness manner in a different set of situations.
- Obtain an overall index of conscientiousness for each individual by averaging the aggregated behavioral measures for each of the twenty situations studied. Correlate this measure with a trait rating derived from self-reports or a person's friends. Can you guess the value of this correlation? It too might well be close to .7. What does this tell us? Perhaps individuals assign trait scores to themselves or their friends by observing characteristic behaviors in many different situations, averaging these observations, and then using this average to reach a judgment about the person. (p. 36)
For an alternative perspective, see The Illusory Nature of Personality Traits.
Activity: Review for the final.
Brody, N., & Ehrlichman, H. (1998). Personality psychology: The science of individuality. Upper Saddle River, NJ: Prentice Hall.