The main purpose of this empirical study was to investigate the predictive effect of school leaders' destructive leadership behavior on teachers' turnover intention. A mixed approach with embedded research design was employed. Multi-stage stratified sampling was applied to select 748 teacher respondents. Data were collected using adapted standardized scale of destructive leadership and teachers' turnover intention instrument. Multiple regression and hierarchical multiple regression model were used to analyze the data. The study revealed that there was a strong, positive correlation between destructive leadership behavior and teachers' turnover intention variables, r = 0.776, n = 748, p < 0.01. It was also founded that except tyrannical, other predictor variables (i.e., derailed, supportive disloyal and laissez faire destructive leadership behaviors) significantly and positively predict teachers' turnover intention, with [beta]=0.04, p>0.05 for tyrannical, [beta]=0.37, p<0.001 for supportive disloyal, [beta]=0.21, p<0.001 for derailed and [beta]=0.30, p<0.001 for laissez fair DLBs. 65% of the variance in teachers' turnover intention was accounted for the interplay of tyrannical, derailed, supportive disloyal and laissez faire destructive leadership behaviors. Supportive disloyal destructive leadership behavior accounted for the highest variation in teachers' turnover intention.