Abstract I analyze the job separation process to learn about gender differences in job separation rates and employment stability. An essential finding is that employer-employee data are required to identify gender differences in job separation probabilities because of labor market segregation. Failure to recognize this may potentially lead to statistical discrimination. Three important empirical results are obtained from the analysis. First, women have higher unconditional job separation probabilities. Second, there are no gender differences in job separation probabilities for employees working in similar workplaces. Finally, women's employment stability is relatively low because they are more likely to move from a job and into unemployment or out of the labor force, and less likely to make job-to-job transitions.