Research for Personnel Attrition

Syndicate content
CONTINUOUS ESTIMATES OF SURVIVAL THROUGH EIGHT YEARS OF SERVICE USING FY 1979 CROSS-SECTIONAL DATA
September 1, 1981
Survival curves for NPS male recruits were estimated through eight years of service using the FY 1979 cross-sectional data base. Separate analyses were performed for Class A school attendees and non-A school attendees, holding constant the effects of age, educational level, and mental group. Mean survival times (the areas under the survival curves) were calculated for each recruit profile. A cost-benefit analysis was then performed on the mean survival times calculated over four years of service to determine optimal qualifying scores for enlistment.
Read More | Download Report
USING QUIT RATES TO SET COMPENSATION LEVELS IN THE PUBLIC SECTOR
June 1, 1981
The government can not base the amount it pays its employees on the market value of the services it provides. Instead, it tries to set pay for each job at a level equal to pay for comparable work in private employment. Although a survey of pay for private sector workers is taken each year, there are indications that the survey is faulty and that government pay levels are too high. One indication is that quit rates are considerably lower in the government than in the private sector. Quit rates are a function of pay levels. The estimating equations in this study are derived from a simple model of labor market behavior.This model makes it clear that efficient behavior for the government does not imply either the same pay or total compensation as the private sector.
Read More | Download Report
MODELS FOR ESTIMATING PREMATURE LOSSES AND RECRUITING DISTRICT PERFORMANCE
February 1, 1981
This paper presents the results of a study with two goals: (1) to develop a model of losses during the first year of service that could be used to evaluate recruiting policy changes and improve the screening of applicants for enlistment; and, (2) to develop a model for evaluating the productivity of Navy Recruiting Districts that could be used in setting quotas, allocating canvassers, and assessing recruiting performance.
Read More | Download Report
A New Look at Success Chances of Recruits Entering the Navy (SCREEN)
June 1, 1980
SCREEN is a table of chances of completing the first year of service used in qualifying applicants for enlistment. The chances are a function of educational level, mental group, age, and dependency status. Because SCREEN was based on regular Navy recruits in 1973, it was updated on recruits and extended to reservists and women who enlisted in 1977. The 1977 regular and reservist SCREENs are compared with one another and with the 1973 version. A women's SCREEN is evaluated in light of current recruiting policy. A streamlined vesion of the 1973 SCREEN is presented, along with a cost-benefit analysis of the qualifying score.
Read More | Download Report
RATING ASSIGNMENTS TO ENHANCE RETENTION
June 1, 1980
The first-term retention gain from exploiting rating-specified survival probabilities when assigning recruits to ratings is assessed. The reassignment of 28,000 recruits to 37 ratings under the same conditions faced in the original assignment is simulated. A sizeable gain in first-term retention rate is demonstrated.
Read More | Download Report
NONPARAMETRIC METHODS FOR ESTIMATING RECRUIT SURVIVAL WITH CROSS SECTIONAL DATA
January 1, 1980
A survey of nonparametric methods (methods which make no distributional assumptions about the data) for survival curve estimation is presented. This is provided as background to the discussion of the Cox regression model, which can be applied to cross-sectional data. The Cox model is then compared to probit analysis on the 1973 recruit cohort of four-year obligors. Evidence is presented to show that the Cox model can be useful for estimating recruit survival from cross-sectional data.
Read More | Download Report
FACTORS THAT AFFECT THE CAREER DECISION OF MARINE CORPS PILOTS AND FLIGHT OFFICERS, VOLUME I, MAIN TEXT AND APPENDIXES A AND B
December 1, 1979
Over the past several years, the Marine Corps has experienced an increasingly high rate of attrition among its pilots and flight officers. Based on a survey of all active duty and some recently separated pilots and flight officers, this study identifies the issues and attitudes that motivate the career decision. Recommendations are made for reducing attrition.
Read More | Download Report
The Quit Rate as a Measure of Job and Pay Comparability
September 1, 1979
The Federal government tries to keep compensation of its civilian employees competitive with compensation in the private sector by setting wages to be comparable to wages in the private sector. Reliance on wages to measure the comparability of compensation may be inferior to reliance on other measures that incorporate pecuniary and non-pecuniary factors not measured by wages alone. Theory suggests that the quit rate could be used to indicate when compensation is getting out of proper adjustment: a rise in the quit rate would indicate that total compensation is falling in comparison with compensation offered by other employees. To judge whether the quit rate is a sensitive measure of compensation comparability, the relation of quits and relative wages in manufacturing industries was examined. Two different data sets were used to test the same general model. Both tests showed that quits rise when relative wages fall, and vice-versa. In one test, using aggregate time series data for each of twenty-six manufacturing industries, all but two industries showed this negative relationship, and the results were statistically significant in twelve of the twenty-six. The other test, using longitudinal earning records of individual workers in the steel industry and in shipbuilding produced results very close to those derived with the more aggregate data. The conclusion is that the quit rate is a sensitive measure of wage comparability. If it is to be used as a basis for adjusting federal wages, however, more must be known a
Read More | Download Report
THE PREDICTION OF ATTRITION FROM MILITARY SERVICE
December 1, 1978
This paper provides a comparison of four statistical models for predicting first-year attrition from the Navy. The models compared are the individual linear probability model, the grouped linear probability model, the individual logit probability model and the grouped logit probability model. For different qualifying scores, the models are compared in terms of their ability to discriminate between attriters and non- attriters. Their ability to predict the actual attrition rates within future entry cohorts is also compared.
Read More | Download Report