This thesis analyzed the effect of military service and military training on post service earnings using the National Longitudinal Survey of young men 14 to 24 years of age in 1966. When data were disaggregated by race and veteran status, some differences appeared between the earnings of veterans and non-veterans. A Chow test indicated that the data could not be pooled for blacks and non-blacks. An analysis of earnings revealed that the effect of veteran status on post-service earnings is inconclusive. Further analysis of wage equations and annual income equations of veterans showed that veteran status does not have significant returns from either military training or time spent in the service. Keywords Statistical tests Data bases Demography.