Correlation analysis was used to analyze what experiences before and during teacher preparation for 72 graduates of an urban teacher education program were associated with urban commitment, first job location, and retention in urban schools for 3 or more years. Binary logistic regression was then used to analyze whether urban K-12 schooling, volunteer service, and student teaching in a high-poverty urban school predicted urban commitment, employment, and retention for at least 3 years in an urban school. The regressions revealed that all three factors predicted strong urban commitment and that urban commitment strongly predicted first job location and retention.The high turnover of teachers in American urban high-poverty schools (defined in this article as those with more than 50% eligible for free/reduced