In-vivo rating of treatment acceptability by children: Effects of probability instruction on student's spelling performance under group contingency conditions

In prior studies, Shapiro and Goldberg (1986, 1990) failed to find a relationship between in-vivo ratings by children of treatment acceptability and treatment effectiveness. These studies involved the use of interdependent and dependent group contingencies designed to improve the spelling performance of sixth grade students. To investigate whether the failure to link treatment acceptability and effectiveness may have been due to the subjects' inability to understand the differences in treatment conditions, this study utilized a pre-intervention training package to enhance salient differences between two types of group contingencies. Results of this study showed that both group contingencies were successful at improving the spelling performance of students, particularly the poorer spellers. Prior to treatment, students preferred the interdependent condition, with the higher-achieving students expressing the strongest preference. After implementation of the training package, both conditions were now rated as equally acceptable. Pre- and post-test acceptability ratings of each condition tended to be significantly correlated but correlations between acceptability ratings and treatment effectiveness were negligible at all points in the study. Limitations of the present study and suggestions for further research are discussed.