Facilitating the acquisition of healthy social attributions in sixth grade students

Three treatments were randomly assigned to 107 sixth grade students. Two classes of students participated in a training program using Rational Emotive Behavior Theory (REBT) and a daily review; the daily review was designed to reinforce the previously taught lessons; two additional classes received the same REBT training program (as the first two classes) but did not participate in a daily review; finally, two classes received a non-REBT social skills training program developed by Sunburst Communications (1995). Training consisted of six lessons over six weel

All participants completed the 30-item Student Social Attribution Scale (SSAS) (Bell & McCallum, 1995) and a 26-item Student Social Success Rating Scale (SSSRS) pre-test and post-test. On the SSAS students selected an attributional dimension/cause (i.e., effort, ability, chance and task difficulty) for failure and success outcomes. Success/Failure-effort plus ability attributions were combined to form the Internal Scale and Success/Failure-chance plus task difficulty, formed the External Scale. On the SSSRS students rated their perceptions of their social success in everyday situations.

Two ANCOVAs compared the adjusted means of the three groups for both dependent variables. The REBT-based training program that includes daily reviews failed to produce more positive attributions for social outcomes than a nonreview-based REBT program (p>.05). Nor were there differences in the students perceptions of their social success (SSSRS mean scores) as a function of training across the three groups (p>.05). Post-test means across the three groups were similar. Similarly, the training program based on REBT failed to produce greater changes in students' development of stable, internal, social attributions and students perceptions for social success than a more conventional ("canned") non-REBT social skills training program (p>.05). However, Total post treatment means were significantly higher than pre treatment for internal attributions (p<.05). The mean pre to post treatment gain for SSSRS scores was in the expected direction, but nonsignificant. Implications are discussed.