US Army's Delayed Entry Program: Attrition Modeling

Abstract

The United States Recruiting Command (USAREC) utilizes the Delayed Entry Program (DEP) as the foundation for their management of the continuous flow of recruits into the training base. Though there are many benefits of the DEP, a major shortcoming is that some DEP members do not enlist, becoming DEP losses. This is costly in terms of valuable resources such as lost recruiter time, and the potential for training seats being unfilled. Any effort which assists in reducing DEP loss would be a valuable contribution. This research models individual level DEP loss using multivariate dichotomous logistic regression. Explanatory variables used were individual, demographic, and USAREC policy in nature. Modeling efforts used data that were easily accessible to USAREC to ensure ease of potential future use. Univariate analysis was conducted on candidate explanatory variables prior to model building. The model was built using forward and backward stepwise logistic regression. Final model refinement included scaling of interval variables and the addition of one interaction term. Using statistical tests, the model as a whole was determined to exhibit some lack of fit. Closer analysis indicated that the model does perform well across many levels of estimated probability of DEP loss. Using USAREC's red, amber, green DEP loss risk classification system, the model appears to have significant predictive powers. The model also performed well using this classification system for a validation data set. It is concluded that this fitted model could prove useful in supplementing the field experience of the recruiter in predicting DEP loss risk of individual recruits.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1991
Accession Number
ADA246288

Entities

People

  • Daniel C. Buning

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Energy and Power Technologies
  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Business Administration
  • Computer Programs
  • Data Science
  • Data Sets
  • Databases
  • Demography
  • Education
  • Geographic Regions
  • Information Retrieval
  • Information Science
  • Military Training
  • Operations Research
  • Probability
  • Social Sciences
  • Statistical Tests
  • Training
  • United States

Readers

  • Computational Modeling and Simulation
  • Naval Personnel Management