Research for CRM

This paper reports on the construction of an interactive, historical database for recruit survival probabilities. This FY 1978-1984 base includes accession and monthly survival information for non-prior service recruits by accession program, gender, shipment mode, educational category, mental group, and age. An important finding is that for recruits who are otherwise identical, survival is worse if they are shipped within the month they signed their initial contract.

Within a few years the Department of Defense may begin administering the Armed Services Vocational Aptitude Battery (ASVAB) using Computerized Adaptive Testing (CAT). In CAT, each test item is characterized by an Item Response Curve (IRC), which describes how the probability of correctly answering the item increases with ability. A recent study conducted by the Center for Naval Analyses found that IRCs of many items in the experimental CAT item pool for the ASVAB changed substantially from paper-pencil to CAT administration. This research memorandum examines the effects of these changes on scores of individual examinees.

Knowing the magnitude of performance differences among individuals is critical in making personnel decisions in the Marine Corps. Recent research at the Center for Naval Analyses has focused on quantifying the magnitude of these performance differences. The purpose of this analysis was to cross-validate those results using hands-on measures of job performance and other published research findings.

Computerized Adaptive Test (CAT) and Paper-and-Pencil (PP) Armed Services Vocational Aptitude Battery (ASVAB) scores are equated in several subpopulations to determine if racial minorities, females, those who did not complete high school, or individuals lacking experience using computers would be at a disadvantage if CAT rather than PP scores were used for selection. Equating is generally independent of population group.