The effects of bibliotherapy on the self-esteem and teacher-rated classroom behavior on third-grade children of divorce
The general purpose of this study was to determine whether bibliotherapy assists students from divorced families in improving self-esteem and classroom behavior. In addition, this study investigated whether a relationship exists between self-esteem, behavior, and gender of students. The subjects in this study consisted of third-grade students from 12 different intact classrooms. The students were categorized as being from either a divorced family or an intact family by information given by the parent or guardian completing a permission sheet. The classrooms were randomly assigned to one of the three treatment conditions (experimental, placebo, and control). The experimental classrooms were read fictional books, by the researcher, dealing with divorce. The readings took place one day a week, for six weeks total. A discussion followed the reading of each book. A different book was used for each session. After the last of the six sessions, the Coppersmith Self-Esteem Inventory (1981) was administered to assess the self-esteem of the subjects, and the teachers completed the ADD-H: Comprehensive Teacher's Rating Scale (2nd ed., 1991). The subjects in the placebo group followed the same procedures described for the experimental group; however, instead of being read a fictional book dealing with divorce, this group was read books written by various award winning authors. Two separate three-way multivariate analyses of variance (MANOVA) (rejection level.05) were used to compare the effects of the three treatment conditions on students' self-esteem and classroom behavior. There were no significant differences among treatment groups on self-esteem ($p\ge.05$), however, a significant main effect for treatments on classroom behavior was found (F (8,256) = 3.15, ($p\le.01$)). There were no significant differences between boys and girls, or divorced and intact families on self-esteem ($p\ge.05$), and no significant difference between boys and girls or divorced and intact families on classroom behavior ($p\ge.05$). No significant interactions were found among treatment, gender and family composition groups for the variable of self-esteem ($p\ge.05$); however, a significant ($p\le.05$) interaction between treatment and family composition on classroom behavior was found. Even though significant differences were not found in self-esteem or classroom behavior through quantitative statistical analysis, the children in the treatment groups showed obvious signs of reaching all three stages of bibliotherapy (identification and projection, abreaction and catharsis, and insight and integration).