An Analysis of the Coast Guard Enlisted Attrition
Abstract
In this thesis, survival analysis is used to study US Coast Guard enlisted attrition behavior in terms of individual personnel characteristics such as sex, marital status, race, paygrade and rating. Results obtained based on 8 years of historical data from FY83 to FY90 are as follows: males and married individuals have higher survival probabilities than their counter parts, respectively; paygrades E-1 to E-5 have higher attrition than paygrades E-6 to E-9; American Indians have the highest attrition and Asian members have the largest survival probabilities; rating 170 (Gunner's Mate) has die highest attrition over all ratings followed by rating 180 (Fire Control Technician); the rating with the highest survival probability is 570 (Aviation Machinist's Mate); a decreasing trend in attrition was found during the last 4 years of the observation period; it was also observed that there was significantly high attrition at the end of the four years service contract and when the enlisted member reaches twenty years of service. Additionally, this thesis provides the a regression model order to predict monthly enlisted attrition figures. Significant predictors selected are the prior month's attrition, the number of enlistments four years prior and the current unemployment rate. The selected regression model explains almost 97% of the total variation of monthly attrition. It turns out to perform better than the current method used by the CG.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1993
- Accession Number
- ADA275139
Entities
People
- Laureano E. Rubiano
Organizations
- Naval Postgraduate School