The purpose of this research is to determine the variables that predict teachers' multicultural education self-efficacy. For this purpose, the Turkey sample data set of the TALIS 2018 Teacher Questionnaire has been discussed. Firstly, the relationships between gender, age, employment status, education level, seniority, first career choice variables and multicultural education self-efficacy were examined by point biserial and Spearman correlation coefficient and t and F tests. Then, the model-data fit of the multicultural education self-efficacy, teaching practices, self-efficacy perception, diversity practices, and teacher collaboration scales in the TALIS 2018 application were determined with the help of confirmatory factor analysis, and their reliability was determined by the Cronbach Alpha and McDonald Omega coefficients. The relationships between the scores obtained from these scales were revealed by the Pearson correlation coefficient, as the scale scores showed a normal distribution. By testing the assumptions of the analysis, multiple regression analysis was performed with the multicultural education self-efficacy variable as the dependent variable, and other scale scores and seniority and first career choice as dummy variables, which were found to be related to the dependent variable, as independent variables. According to the results of the analysis, it was found that there are low but significant relationships between age, seniority, first career choice and teachers' multicultural education self-efficacy and, there are no significant relationship between gender, employment status, education level and multicultural education self-efficacy. It was determined that teachers' multicultural education self-efficacy and teaching practices, self-efficacy perception, diversity practices, teacher cooperation variables show moderate significant relationships. As a result of the multiple regression analysis, it was concluded that the strongest predictors of teachers' multicultural education self-efficacy were teacher self-efficacy, teaching practices, diversity practices, seniority, and teacher collaboration, respectively. It was observed that these variables together could explain only %22 of the variance in the dependent variable.