A STATISTICAL MODEL FOR THE PREDICTION OF REENLISTMENT.
RESEARCH ANALYSIS CORP MCLEAN VA
Pagination or Media Count:
The paper deals with the development of Army reenlistment rates by relating specific causal factors through multiple regression analysis. With FY66 separation records as a data base, investigation revealed that grade, race, military occupational specialty MOS, and state of residence were the most significant factors in explaining reenlistment decisions. These were combined as parameters in a regression equation that provides an improved basis for estimating the number of men who will reenlist at the expiration of their term of service. Several tables of reenlistment rates are presented in the paper, but these are based on FY66 data only. The most important results of the study are not the actual rates but the exposition of a methodology and a model that improve the capability for prediction. The model parameters also provide the basis for analysis of important manpower policy variables. Author
- Humanities and History