Research training environment (RTE) has long standing in the psychological field, which improved the research interest in students, yet the study of RTE in the educational field still limited. Was the RTE model with the research training environment and research interest mediated by research outcome expectations and research self-efficacy fitted to the data well? The present study were 1) to develop and validate a causal model of RTE and 2) to test the mediating effects of research outcome expectations and research self-efficacy between research training environment and research interest. The questionnaire were translated and adapted from previous studies. Samples of the study consisted of 141 graduate students in the faculty of education, a national research university of Thailand. Structural equation modeling with maximum likelihood estimation was used to validate the research training environment model. Various tests were employed to examine the indirect effects in the model. The model was fitted to the data well (chi square = 24.292, df = 19, p = 0.185, CFI = 0.989, TLI = 0.978, RMR = 0.042, RMSEA = 0.045). Indirect effect testings indicated that research self-efficacy and research outcome expectations mediated the relationship between the research training environment and research interest. Even though the research training environment has no direct effect on research interest, it still is an important factor in the way of developing an interest of graduate students in the educational field as well as psychological related field. Besides, research training environment is such a practical factor that appropriate intervention can be developed.