Doctor of Philosophy (PhD)


Social Work

Document Type



The major focus of this study was to examine how children’s perceptions of attachment security and children’s emotion regulation predict school disengagement among elementary school truants. This study utilized an exploratory, cross-sectional survey design. Data collection was accomplished with four standardized survey instruments, completed by 74 truants (ages 7 to 12 years) and their parents and teachers, and secondary data collected by the East Baton Rouge (EBR) Truancy Assessment and Services Center (TASC), Louisiana during the 2006-2007 academic years. The school disengagement model was tested with three latent variables (children’s perception of emotional bonds, emotion regulation, and school disengagement) using structural equation modeling (SEM). Also, this study tested a multiple indicators-multiple causes (MIMIC) model in the SEM to examine the relationship between three sample characteristic variables (age, grade, and resistant status) and three latent variables. Additionally, hierarchical regression analysis was utilized to support the school disengagement model. The structural equation model had a good fit to the data (AGFI = .88, NFI = .91, RMSEA = .00). The findings indicate that children’s emotional regulatory capacity is a significant predictor of school disengagement. Contrary to expectations, the emotional bonds of children did not significantly predict emotion regulation or school disengagement in the model. However, bi-variate results indicated that two indicators of emotional bonds, children’s perceptions of attachment security and trust, were associated with children’s risk for aggressive behavior. The result of hierarchical regression also showed that children’ emotional bonds was statistically significant predictor of children’s externalizing behavior problems at .10 level. The MIMIC model showed that gender was a good predictor of children’s perceptions of attachment security, but age and resistant status were not predictors on all latent variables in the model. This study extends the attachment and emotion regulation research with an examination of school disengagement in high-risk, middle-childhood, African American children. The findings are discussed with regard to the implications of the prediction of school disengagement from emotional bonds, emotion regulation, and gender differences. The findings concerning emotion regulation are also discussed in relation to their implications for truancy intervention programs to help high-risk elementary school children.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Timothy Page



Included in

Social Work Commons