Turnover of employed bus operators increases cost of service for transit agencies, meaning interventions to reduce turnover can be a common goal. To be effective, interventions should be based on an understanding of the types, timelines, and underlying reasoning for operators separating from employment. However, to date there has been little quantitative research into bus operator separation dynamics. To better understand these dynamics, a hierarchical survival model was used to quantify monthly probability of separation for individual bus operators, based on a sample of two-year work histories for over 1300 bus operators employed by Metro Transit (Minneapolis-St. Paul, MN). Characteristics of operators at hire, such as age, gender, and driving experience, as well as work history events such as absences, incidents, and accidents, were used to understand what influences separation. Separation rate differed predictably with time, with the highest risk of both voluntary and involuntary separation in the first ten months of operator employment. Of bus operator characteristics at hire, prior commercial driving experience lowered the probability of separation. The number of absences from work, and accidents classified as the responsibility of the operator, increased the separation risk probability per month. The highest probability of separation was in operators hired without a commercial driver's license, whose work history included multiple accidents. Alterations to hiring and training practices may ameliorate risk of bus operator separation. Additionally, the hierarchical modeling approach utilized here has promise for predicting relative impacts of characteristics of employees and their work history on multiple aspects of employment.Effective public transportation relies in great part on the everyday labor of bus operators, who can fulfill the scheduled service safely and on time. The job demands are unique: one must be able to operate a large vehicle in mixed traffic while adhering to a schedule, monitoring electronic communications, and providing retail-level interaction with customers. Because of these demands, not every employee who begins as a part-time bus operator will spend a career as a transit employee; some will leave voluntarily, and a smaller fraction will be involuntarily separated from employment. These separations are costly to transit agencies, as they include not only the sunk costs of recruitment and training of the departing employee, but also lost productivity during an interim period before the employee can be replaced (1).Because of the high cost of bus operator turnover, transit agencies have an incentive to understand the factors that might predict operator success and retention. Previous research has focused mainly on differentiating potential operators at the time of hire. For instance, Jacobs et al. (2) surveyed bus drivers across nine organizations using a battery of over 350 questions, which were then combined into scores and compared to supervisory ratings, absences, and accident data...