Prediction of Navy E-4 Test Passers.
NAVAL POSTGRADUATE SCHOOL MONTEREY CA
Pagination or Media Count:
This thesis applies hierarchical clustering and quadratic discriminant function techniques to the problem of predicting E-4 test passers and non-test passers including non-test takers in the Navy. The biographic data base includes items such as test scores and education to serve as separators and predictors in the techniques. The clustering of rates permitted accumulation of personnel in the lightly staffed ratings into similar groups of substantial size, was objective, and may be useful for purposes other than the present one. The discriminant analysis produced correct classification rates of about 60 to 70 percent based on the data at hand. Author
- Statistics and Probability