Assessing Instructor Effectiveness Based on Future Student Performance

Educational institutions rely on instructor assessment to determine course assignments, which instructors to retain or promote, and whom to provide with additional assistance or training. Instructor assessment is most commonly based on student surveys or peer evaluation— which are both subject to the evaluator's personal biases. This study describes an assessment method based on future student grade performance, which has the potential to avoid these biases. This study is based on eight years of undergraduate course-grade data from over 24,000 students in a large metropolitan university. The methodology introduced in this paper accounts for confounding factors, such as diverse instructor grading policies and varying student abilities. Top and bottom performing instructors are identified for each course.