Effects of controllable versus uncontrollable factors on responsibility attributions: A single-subject approach.

Abstract

A methodology was used to assess the influence of different informational cues on responsibilit y attributions for an automobile accident. Multiple regression analyses indicated that although there were large individual differences, in Experiments 1 and 2 the major determinants of responsibilit y judgments were car speed and the condition of the car's brakes. In Experiment 3, speeding and brakes information was deleted. The major determinant of attributions was the driver's past record, although the importance of severity of consequences was somewhat greater than in the first two experiments. In Experiment 4, driving record was deleted. The majority of subjects indicated that not enough information was available. Overall, severity was a relatively unimportant cue. Observers tended, instead, to rely on factors over which the driver had control. The lens-model approach seems quite suitable for studying attributional judgments, particularly since it provides a detailed description of individual judgment strategies. Attribution research has typically utilized experimental designs in which only a limited number of attributional determinants are manipulated. Additionally, between-subjects statistical analyses are used in which individual differences are treated as error variance. The present studies were conducted to explore an alternative approach that permits the assessment of the effects of several independent variables and the determination of the relative importance of each variable for individual subjects. Fischhoff (1976) has pointed out that attribution theory is not the only area of psychological research concerned with inferential behavior. An attribution can be conceptualized as a special case of a general judgment process wherein individuals integrate information to arrive at judgments.