Abstract: This research aimed to investigate academic ethics in research methodology of Mahasarakham University’s Graduate Students, to compare between academic ethics in research methodology of Masters students and that of doctoral students and to construct predictive equations of factors that were related to behavioral academic ethics in research methodology. The samples of the study were 66 doctoral students and 434 Master students, who were enrolled in the faculty of Education, Mahasarakham University. The research instrument included: a 30-items scale on the academic ethics in research methodology with discriminating powers ranging from 0.19-0.63 and a reliability of 0.8645. The collected data were analyzed by percentage, mean and standard deviation t-test. Pearson Product-Moment Correlation Coefficient and Multiple Regression Analysis were employed for hypothesis testing. The results of the study showed: that all students who involved as the sample of the study had average scores of 26.35 (SD = 3.00) on academic ethics knowledge and 26.41 (SD = 3.06) on academic ethics behavior, while their attitude score toward academic ethics was at a high level, that there was no difference among masters students and doctoral student in terms of academic ethics knowledge score, academic ethics attitude score and academic ethics behavior score and that academic ethics knowledge and attitude toward academic ethics could predict academic ethics behavior. The multiple correlation coefficient (R) was 0.908 with predicting powers at 82.40%. The equations in the raw-scores form and standard-score form could be written as: the predictive equations in the raw-score form: Y’ = 1.072 + 0.877 know + 0.560 Att and the predictive equations in the standard-score form: Z_y’ = 0.860Z_know + 0.081Z_Att.