Using an opportunity-propensity framework to estimate individual-, classroom-, and school-level predictors of middle school science achievement

Using data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (N = 4447), this analysis employs an opportunity-propensity (O-P) framework (Byrnes, 2003; Byrnes & Miller, 2007; Byrnes & Wasik 2009) to examine the influence of multiple student, teacher, classroom, and school factors on eighth-grade science achievement. Saçkes, Trundle, Bell, and O'Connell (2011) fit an O-P structural equation model (SEM) to the same database to explain science achievement growth from Kindergarten to third grade. We extend this work by fitting an O-P SEM to this database to predict science achievement growth from fifth to eighth grade. This middle school model includes an opportunity variable - science curriculum track placement - that operates only in middle and high school. This variable and the school's poverty rate are significant predictors of several opportunity factors. We replicate previous findings that propensity factors are the strongest determinants of science achievement, notably prior achievement. However, we find more opportunity factors than previous studies that are also significant. Other things being equal, having a state-certified teacher is the second strongest predictor of achievement within the model. Placement in a science honors course and being enrolled in a low income school are also linked to small but significant impacts on science achievement.