Teacher turnover continues to be a significant problem in the United States. Teacher turnover is expensive because it costs money to continue recruiting, hiring, and training new teachers to replace those leaving (Carver-Thomas & Darling-Hammond, 2017). Most important though, teacher turnover hurts student achievement and success (Sorensen & Ladd, 2020). This study aimed to investigate ways data mining could be used to predict teacher turnover through building machine learning models. Machine learning algorithms and other data mining techniques are underutilized in education, especially regarding teacher retention (Plotnikova et al., 2020). This secondary, quantitative study was conducted by building five machine learning models using public-use data from the National Center for Education Statistics (NCES). Decision tree classifiers, gradient boosted trees, random forest trees, logistic regression, and support vector machine models were built using the School and Staffing Survey (SASS) and the Teacher Follow-Up Survey (TFS) initially collected by the NCES. The models were then analyzed for accuracy, variable correlation, and significance. The study found that predictive models can quickly and accurately be used to predict teacher turnover and factors that most affect teacher turnover using large datasets. The model's accuracy and representation of the data set allow for generalizability. The findings from this study could help practitioners and policymakers make better-informed educational decisions. This study provides the field of education with an additional tool to combat the problem of teacher turnover. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]