Causal Attributions, Attributional Dimensions, and Academic Performance in a School Setting

The attribution model of achievement motivation has been applied to academic achievement as a way of understanding underachievement and as a basis for developing intervention programs. There has been little applied research in this area, however, that supports the use of the model in school settings. The purpose of the present study was to test the applicability of the model to an actual school setting. Subjects were 149 tenth grade students in a large urban school district. In accordance with the model, specific attributions for success or failure were assessed, as well as subjects' perceptions of the locus, stability, and controllability of attributions. Attribution patterns found in previous analog research were not found in a school setting. Immediate effort attributions were the most prevalent, regardless of performance level or outcome. Causal beliefs were found to relate to performance in ways predicted by the model but also in some ways not predicted. Relationships were generally stronger for high performers. Comparing subjects' perceptions of the dimensional properties of attributions across outcomes showed a strong outcome bias. Attributions were perceived as more internal and stable following successes, consistent with previous research. In addition, a performance level bias was found. Low performers rated attributions as less internal, stable, and controllable following successes and more so following failures than did high performers. This bias, termed the underachievement bias, was discussed in terms of its detrimental effects on school performance. The differences between high and low performers regarding perceptions of dimensionalities were consistent with the predictions of the attribution model. It was concluded that the attribution model is applicable to school settings. Suggestions were made that more applied research be conducted, that intervention programs based on this model should target subjects' perceptions of attributions rather than just the specific attributions themselves, and that because of the differences among subjects in perceptions of dimensional properties of attributions, researchers should obtain a direct measure of subjects' perceptions rather than assuming them.