Attribution

Attribution Theory

Attribution theory has its roots in Fritz Heider's view, published in 1958, that human beings are naive or intuitive psychologists who are in the business of understanding their social world through a process of cause–effect analysis. That is, people, very much like formally trained psychologists, seek to attribute causes to behavior. Clearly, if you know the causes of behavior, then you can anticipate behaviors and you can choose to engage in behaviors that you know have certain outcomes. Attribution renders the world predictable and controllable—a place within which people can act effectively and adaptively.

This general idea was explored most fully in the work of Harold Kelley in the late 1960s and early 1970s. Kelley believed that people are extremely rational in determining the causes of behavior. They cognitively compute, much like a statistician, [Page 49]the degree of covariation/correlation between a behavior and a potential cause of the behavior in order to decide whether the potential cause is the actual cause or not. In determining covariation, people draw on three sources of information: consistency information, distinctiveness information, and consensus information. In attributing causes to behavior, people are particularly interested in knowing whether a behavior reflects an enduring internal property of the person (personality) or an external property of the particular situation the person finds himself or herself in.

This is most easily explained with an example. Imagine that you meet someone, let's call her Jess, who is friendly toward you. What does her friendliness tell you? If she is only sometimes friendly toward you, her behavior is low in consistency—her friendliness is not due to you, and it does not reflect a general tendency on her part to be friendly. Under these circumstances, you would discount yourself or her as a cause of her friendliness, and you would look for other potential causes. If, however, she is always friendly toward you, then her behavior is consistent—but further information, on distinctiveness and consensus, is needed to know whether the cause of the friendliness is her personality or something about the situation she is in (for example, with you). If she is always friendly to you but not to anyone else, and not everyone is friendly to you, then her behavior is high in distinctiveness and low in consensus, and you can probably assume that she specifically likes you. If, on the other hand, Jess is not only friendly to you but to everyone else, and most other people are not friendly toward you, then distinctiveness and consensus are both low, and you can assume that Jess is simply a friendly person.

...