Many theoretical models have been suggested; they usually involve a search for the cause of the behaviour, and an attempt to determine whether this is internal or external, followed by a decision as to whether the behaviour is logical or rational. Some of the most prominent theories are initially Heiders (1948) Model, which highlighted the fact that all behaviours have both internal and external causes. There is also Jones and Davis (1965) Correspondent Inference Theory that is concerned with how we move from observing behaviour, to understanding the intention of the actor, to concluding the disposition which caused the intention.
There is also Kelleys (1967) ANOVA model, which is concerned with what information we use to arrive at a causal attribution. In this essay, I shall attempt to further discuss and evaluate Kelleys model, as it is one of the most recent and widely discussed. Inevitably, all of these theories have been further formalized and extended by later psychologists. The key question for Kelley (1967) was what type of information does the person use in order to make a causal attribution? (Scott and Spencer, 1998). He saw the person as a nai??
ve scientist who weighs up several pieces of information before arriving at an explanation for events causes. Kelly believes that in order to determine the origins of a behaviour, for example Nicola argued with her dad, we need three types of information; distinctiveness, consensus and consistency. Distinctiveness refers to information we require about the stimulus, which in this case is Nicolas dad. If Nicola only argues with her dad the distinctiveness is high, however, if she argues with everyone the distinctiveness, of her dad, is low. The second type of information that is needed is consensus details about Nicola herself.
There is high consensus for Nicolas argumentative behaviour if many people argue with her dad and low consensus if it is only she. Finally, consistency information, about the circumstances surrounding the argument, needs to be considered. For example, there is high consistency if Nicola argues with her dad in many situations, but low consistency if Nicola has only argued with her father on this specific occasion. When all of this information has been obtained we use the principle of covariation, deciding what factor is always present with this behaviour, to determine what was the cause of the argument.