Student and Teacher-Level Predictors of Advanced Placement Physics Performance

  The question of Advanced Placement (AP) Physics achievement has been a persistent concern when considering the goal of diversifying participation in post-secondary physics study, since students who perform well have academic advantages. This observational study examined demographic and teacher-level predictors of AP Physics 1 performance in the U.S. Potential variables from theoretical models included student ethnicity, socioeconomic status, and gender, as well as teacher experience, certification, course load, and gender. Correlated predictors were included in a multivariable model to understand potential constraints in the STEM pipeline. Descriptive and inferential analyses of purposefully sampled state-level data (N=102 high schools, 2607 students) were conducted. On average, women accounted for 38% of a school's test-taking population, underrepresented minority students accounted for 13.5%, and economically disadvantaged students 17.2%. On average, 60% of students scored a 3 or higher on the examination, the average numerical score was a 2.88 out of a possible 5. Teachers of AP Physics tended to be certified in the subject, taught physics exclusively, and averaged 18.5 years of teaching experience. Multivariable linear regression indicated AP Physics performance was predicted by the socioeconomic status and gender of the test-takers (F(3,97)=12.191, p<0.001), with men outperforming women. These two student demographic variables accounted for 27.2% of the variance in scores. Teacher-level variables were not significant predictors of performance, including certification type, professional age, and gender. Results suggest that women and students of low socioeconomic status, who have been traditionally marginalized in the physical sciences, may require interventions to boost their AP Physics performance. School leaders and policy makers should implement more proactive interventions to promote diverse physics participation and more equitable performance outcomes.