Attribution theory, which focuses on student beliefs about why they succeed or fail, was investigated with 220 sixth grade students who were required to pass the Virginia Literacy Passport test. Part I of this study investigated the reliability and validity of Causal Dimension Scale II (McAuley et al., 1992) when it is used with a sixth grade preadolescent population. Part II investigated student attributions for performance outcomes on the reading portion of the Virginia Literacy Passport test. This study researched the hypothesis that middle school students who passed the Virginia Literacy Passport Test (LPT) reading test would attribute their successful outcome to more internal, stable and controllable causes than students who did not pass the test, Reliability coefficients for CDSII were within Nunnally's (1967) acceptable range for new instruments. Factor analysis supported Russell's hypothesized two factor structure of personal and external control. However, in the main study, three instead of four factors emerged with personal control and locus appearing to measure similar constructs. As predicted, t-tests results indicated that the students who passed attributed their outcome more to internal, stable, and controllable causes. ANOVA results indicated no significant differences between gender and ethnicity, however, there was a significant interaction between minority and nonminority males and females. Curriculum implications include affective education and attribution retraining for the students and their parents. Urban policy implications include cross-curriculum reading instruction, staff development initiatives to educate teachers about the role of causal attributions in motivating students, and parental and community involvement in reading enrichment programs.