Abstract
The purpose ef this study is to examine Aronson & Carlsmith's hypothesis concerning risk taking behavior from the point of view of achievement motivation, using anagram solution task. Used as subjects were 47 high school girls. As independent variable, expectancy and performance were operated, and four (2×2) experimental conditions were provided. The hypotheses were as follows. (1) In risk taking behavior, a high expectancy-high performance condition is the riskiest, and a low expectancy-high performance condition comes next. A high expectancy-low performance condition, along with low expectancy-low performance condition, may be conservative, but the former may be riskier than the latter. (2) Therefore, in actual performance after risk taking, a high expectancy-high performance condition comes first, followed by a low expectancy-high performance condition and high expectancy-low performance condition in the order mentioned, with a low expectancy-low performance condition at the bottom.
Analysis of variance showed that performance (p<. 01) affects risk taking behavior more than expectancy (p<. 025) does, and that the interaction is not significant. This result indicates that the Aronson & Carlsmith's hypothesis is not applicable to risk taking behavior. Which means that the hypothesis (1) was supported. It was found that in those groups in which expectancy and performance coincide, the effects of expectancy and performance combine to affect risk taking behavior, whereas in those groups in which expectancy and performance do not coincide, risk taking behavior was dominated mainly by successful experiences, irrespective of expectancy and performance.
Futhermore, no significant relationship was found between performance expectancy and actual performance, and between risk taking behavior and actual performance. Thus, the hypothesis (2) was not supported.
