Factors influencing college teaching self-efficacy of science, technology, engineering, and mathematics graduate teaching assistants

Ceylandağ, Rana
This study aimed to test a statistical model among GTA teaching self-efficacy, instructor’s teaching support, departmental teaching support, teaching experience of GTAs, number of courses they assisted, mastery experiences, and value they attribute to college teaching through structural equation modeling. Data were collected from 302 GTAs (159 female, 142 male) studying science, technology, engineering, and mathematics (STEM) disciplines with a response rate of 30.82%. In order to collect data, GTA Survey Instrument was developed and this instrument included following sections: GTAs’ demographic information and teaching profile, mastery experience, value attributed to college teaching, teaching self-efficacy, teaching experience, instructor’s teaching support, and departmental teaching support. In this study, DeChenne (2010)’s GTA Teaching Self-Efficacy Scale was adapted to Turkish in order to measure GTA teaching self-efficacy. Current study provided valid and reliable findings for the adapted version of the scale. In addition, other sections also showed satifactory psychometric characteristics. Analysis resulted in a moderate fit with the following indices: χ2 (1449, n = 302) = 2351.90, p = .00, CFI = .91, NNFI = .90, RMSEA = .05, and SRMR = .08. Mastery experience and value attributed to college teaching were found as significant predictors of GTA teaching self-efficacy. Departmental teaching support had an influence on teaching self-efficacy through value, whereas teaching approach and teaching support of instructor were found as non significant in estimating teaching self-efficacy. Mastery experience was also significantly predicted by number of courses assisted and teaching experience. The model explained 72% of variance in GTA teaching self-efficacy.