Predicting Preservice Teachers Performance on the Science Core of the EC-6 TExES General Certification Examination

Abstract

Predicting preservice teachers’ performance on their certification examination may meaningfully help Educators Preparation Programs (EPPs) to adapt and integrate learning frameworks that can improve their passing rates. This study used multiple linear regression (MLR) and binomial logistic regression (BLR) to explore potential variables that may impact the preparedness of 170 pre-service teachers to pass the science core of the official EC-6 Texas Examinations for Educator Standards (TExES) certification examination. The study was conducted by issuing a practice EC-6 TExES certification examination in a pretest and post-test manner during the semester that the participating cohort were enrolled in BIOL 1082, a mandatory science course EC-6 preservice teachers need to take prior to the official state EC-6 TExES certification exam. Additionally, the cohort took an online Qualtrics^TM survey that collected ex post facto and other demographics data. The independent variables explored in this study included: final grade in BIOL 1082, classification, transfer status, prior college science coursework, enrollment status, family’s college history, and current GPA. The dependent variable used was the post-test score on the practice EC-6 exam. The independent variable, grade in BIOL 1082 was revealed to be the single best predictor of preservice teachers’ performance on the science practice examination across both the MLR and BLR models. The BLR models had a higher prediction accuracy of preservice teachers who would most likely fail the practice test than those who may pass at a prediction rate at approximately 79% accuracy. Based on the 67 out of 170 preservice teachers who passed the post-test, the accuracy of predicting failures may be a useful tool that EPPs can use in identifying students who may be at risk of failing and thus implement necessary interventions and other educational strategies.