A Statistical analysis of officer retention in the U. S. Military

This thesis examines the effect of officer commissioning sources on the retention of officers at two different career points: (1) At the end of the initial service obligation (MSR), and (2) at ten-years of service. The goal of this study is to help policymakers in setting and implementing personnel policies by providing information on the effectiveness of each commissioning program. The Defense Manpower Data Center in Monterey, California, provided the data file used in the analysis of officer retention. The data file contained longitudinal information on the population of officers who entered the military between 1985 and 1995. Logit regression models were used to analyze officer retention at MSR and at the ten-year point. Results indicate that retention behavior varies across commissioning programs. Significant differences in retention are observed among graduates of the Service Academies, ROTC Scholarship and ROTC Non-scholarship Programs, Officer Candidate/Training Schools, and Direct Appointment Programs. The differences are observed for all services combined and for each individual service. In most of the models, commissioning source variables are significant; however, the magnitude differences in retention between the five major commissioning sources often are not large. Moreover, the direction of the retention effect often varies across the services for each commissioning program. Further research on officer commissioning programs is recommended to include individual preferences and job satisfaction in the analysis of officer retention.