US Army Post Initial Entry Training First-Term Attrition Analysis: Part 2

Abstract

This research continues previous work in Army post Initial Entry Training (IET) first-term attrition by including medical data with existing personnel data to predict and understand attrition. We use supervised machine learning models to (1) identify the demographic and medical factors of Army enlisted personnel with highest probability of failure to implement preventative measures and (2) estimate total failures during the first enlistment term to set proper recruiting targets. We use classification and survival analysis techniques within the Person-event Data Environment (PDE) to inform sponsors on attrition trends. We use model results as inputs to an application that displays the predicted probability of success for first term enlistees. The results and application have applicability to other DoD organizations concerned with accession and retention. We find that a soldiers medical history, particularly his Dental Class, PULHES Deployable status, and the duration of the initial contract are significant predictors of whether a soldier will complete his or her first term. Knowledge of the key factors and other influencing variables assists the Army Resiliency Directorate in creation of models to better advise U.S. Army leadership on intervention strategies and preventative measures to preclude the loss of first-term soldiers.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 06, 2019
Accession Number
AD1088669

Entities

People

  • Aaron Devig
  • Anthony D. Smith
  • Gabe Gobea
  • Lyn R. Whitaker
  • Samuel E. Buttrey
  • Ta'lena Fletcher

Tags

Communities of Interest

  • Biomedical
  • Human Systems

DTIC Thesaurus Topics

  • Army Personnel
  • Attrition
  • Data Science
  • Education
  • Enlisted Personnel
  • Information Processing
  • Information Science
  • Machine Learning
  • Management Personnel
  • Medical Personnel
  • Military Personnel
  • Organizational Structure
  • Pain
  • Personnel Management
  • Recruiting
  • Supervised Machine Learning
  • Training

Readers

  • Naval Personnel Management
  • Systems Analysis and Design
  • Trauma or Military Medicine

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks