Predicting Teacher Value-Added Results in Non-Tested Subjects Based on Confounding Variables: A Multinomial Logistic Regression

Teacher value-added measures (VAM) are designed to provide information regarding teachers’ causal impact on the academic growth of students while controlling for exogenous variables. While some researchers contend VAMs successfully and authentically measure teacher causality on learning, others suggest VAMs cannot adequately control for exogenous influences on the classroom. Furthermore, because VAMs are primarily connected to student performance on standardized, high-stakes exams and those exams are resoundingly considered to be inadequate measures of true student learning, educators and educational leaders assert VAM results are moot. The purpose of this study was to consider the potential for student background, teacher preparation, and school climate variables to predict teacher VA classifications in arts education courses, health and physical education, and world languages. Participants were drawn from a sample population of teachers representing (n = 84) elementary, (n = 44) middle school, and (n = 61) high school teachers from an urban North Carolina school district. Data were collected from the North Carolina Teacher Working Conditions Survey (NCTWCS), the North Carolina Analysis of Student Work (ASW), and archived data made accessible by the school district. A multinomial logistic regression (MLR) was conducted to analyze the predictive potential of exogenous variables on VA categorical classification in the absence of a standardized, high-stakes exam. Results for ASW 15 were ultimately not significant; however, results for ASW 16 indicated teacher licensure, professional development, and experience were compounding variables. Recommendations for future research include conducting a path analysis to ascertain effects of combining various exogenous variables on VA classification