Learning to Teach Effectively

Abstract

Graduate teaching assistants (GTAs) from science, technology, engineering, and mathematics (STEM) are important in the teaching of undergraduate students (Golde & Dore, 2001). However, they are often poorly prepared for teaching (Luft, Kurdziel, Roehrig, & Turner, 2004). This dissertation addresses teaching effectiveness in three related manuscripts: 1. A position paper that summarizes the current research on and develops a model of GTA teaching effectiveness. 2. An adaptation and validation of two instruments; GTA perception of teaching training and STEM GTA teaching self-efficacy. 3. A model test of factors that predict STEM GTA teaching self-efficacy. Together these three papers address key questions in the understanding of teaching effectiveness in STEM GTAs including: (a) What is our current knowledge of factors that affect the teaching effectiveness of GTAs? (b) Given that teaching self-efficacy is strongly linked to teaching performance, how can we measure STEM GTAs teachingself-efficacy? (c) Is there a better way to measure GTA teaching training than currently exists? (d) What factors predict STEM GTA teaching self-efficacy? An original model for GTA teaching effectiveness was developed from a thorough search of the GTA teaching literature. The two instruments – perception of training and teaching self-efficacy – were tested through self-report surveys using STEM GTAs from six different universities including Oregon State University (OSU). The data was analyzed using exploratory and confirmatory factor analysis. Using GTAs from the OSU colleges of science and engineering, the model of sources of STEM GTA teaching self-efficacy was tested by administering self-report surveys and analyzed by using OLS regression analysis. Language and cultural proficiency, departmental teaching climate, teaching self- efficacy, GTA training, and teaching experience affect GTA teaching effectiveness. GTA teaching self-efficacy is a second-order factor combined from self-efficacy for instructional strategies and a positive learning environment. It is correlated to GTA perception of teaching training and university GTA training. The K-12 teaching experience, GTA perception of teaching training, and facilitating factors in the departmental climate predict STEM GTA teaching self-efficacy. Hours of GTA training and supervision are fully mediated by perception of GTA training. Implications for research and training of STEM GTAs are discussed.