Longitudinal predictors of high school completion

Area of Science:

• Educational Psychology
• Developmental Psychology
• Child Psychiatry

Background:

• Student dropout is a significant concern in educational systems.
• Identifying early predictors is crucial for timely intervention.
• Existing research highlights various demographic and academic factors associated with dropout.

Purpose of the Study:

• To longitudinally examine predictors of elementary school dropout.
• To develop and compare predictive models based on student characteristics.
• To identify early warning indicators for student completion and dropout.

Main Methods:

• Longitudinal study design.
• Utilized student demographic data, academic achievement (Iowa Tests of Basic Skills), attendance records, and behavioral ratings (Behavior Assessment System for Children).
• Developed two predictive models (Model A and Model B) varying on student sex and race to predict dropout and completion.

Main Results:

• Two models predicted dropout with 75% to 88% accuracy.
• Model A (Behavioral Symptoms Index, School Problems, Iowa Tests of Basic Skills, work habits) best predicted female and African American dropouts.
• Model B (Adaptive Skills, Externalizing composite, School Problems, student support team referral, sex) was more accurate for Caucasian dropouts.
• Both models had similar prediction rates for male dropouts.

Conclusions:

• Behavioral and academic indicators in elementary school are strong predictors of dropout.
• Predictive models vary in accuracy based on student demographics (sex and race).
• Findings support the development of early warning systems and targeted interventions for at-risk students.