Longitudinal predictors of high school completion.

This longitudinal study examined predictors of dropout assessed in elementary school. Student demographic data, achievement, attendance, and ratings of behavior from the Behavior Assessment System for Children were used to predict dropout and completion. Two models, which varied on student sex and race, predicted dropout at rates ranging from 75% to 88%. Model A, which included the Behavioral Symptoms Index, School Problems composite, Iowa Tests of Basic Skills battery, and teacher ratings of student work habits, best predicted female and African American dropouts. Model B, which comprised the Adaptive Skills composite, the Externalizing composite, the School Problems composite, referral for a student support team meeting, and sex, was more accurate for predicting Caucasian dropouts. Both models demonstrated the same hit rates for predicting male dropouts. Recommendations for early warning indicators and linking predictors with interventions are discussed.

Measuring student engagement among elementary students: pilot of the Student Engagement Instrument--Elementary Version.

Early school withdrawal, commonly referred to as dropout, is associated with a plethora of negative outcomes for students, schools, and society. Student engagement, however, presents as a promising theoretical model and cornerstone of school completion interventions. The purpose of the present study was to validate the Student Engagement Instrument-Elementary Version (SEI-E). The psychometric properties of this measure were assessed based on the responses of an ethnically diverse sample of 1,943 students from an urban locale. Exploratory and confirmatory factor analyses indicated that the 4-factor model of student engagement provided the best fit for the current data, which is divergent from previous SEI studies suggesting 5- and 6-factor models. Discussion and implications of these findings are presented in the context of student engagement and dropout prevention.

Curriculum-based measurement oral reading as an indicator of reading achievement: a meta-analysis of the correlational evidence.

This meta-analysis summarized the correlational evidence of the association between the CBM Oral Reading measure (R-CBM) and other standardized measures of reading achievement for students in grades 1-6. Potential moderating variables were also examined (source of criterion test, administration format, grade level, length of time, and type of reading subtest score). Results indicated a significant, strong overall correlation among R-CBM and other standardized tests of reading achievement and differences in correlations as a function of source of test, administration format, and reading subtest type. No differences in the magnitude of correlations were found across grade levels. In addition, there was minimal evidence of publication bias. Results are discussed in terms of existing literature and directions for future research.