Empirical Weighting of Predictors for the Naval Academy Selection Program.
Abstract
A selection composite, the Candidate Multiple, was studied for revision to find an optimal combination of present and experimental predictors. Data from three Naval Academy classes (1975-1977) were used to determine and cross validate regression weights that best predict the following criteria of success at the Academy: (1) Cumulative quality point ratio (CQPR), (2) Cumulative military aptitude rating (CMAR), (3) Choice of major (MAJOR), (4) Voluntary resignation (VR), and (5) All attrition (ATT). Zero-order validities were computed against each of the five criteria for each of seven predictors. Two composites of almost equal value were developed that exceeded the prediction obtained with the current composite.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1976
- Accession Number
- ADA027275
Entities
People
- Idell Neumann
- Norman M. Abrahams
Organizations
- Bureau of Naval Personnel