Research for Aptitude

Hands-on performance tests are the benchmark against which other measures of proficiency should be compared. However, hands-on performance tests are expensive, time consuming, and sometimes dangerous to personnel or equipment. This paper analyzes the relationship between hands-on performance tests and job-knowledge tests. The study extends previous studies to indicate the conditions under which job-knowledge tests should be used as surrogates and to provide guidance for developing better job-knowledge tests.

This research memorandum presents and applies a methodology for estimating the cost of recruiting individuals with alternative distributions of Armed Forces Qualification Test (AFQT) scores. The methodology takes account of the key institutional features of the recruiting process, including recruiter time allocation and procedural guidelines. The method is used to estimate the costs of different recruit-aptitude distributions, using data on applicants and accessions for all of the services.

The Numerical Operations (NO) subtest has caused problems for the joint-service testing program since it was made part of the Armed Forces Qualification Test (AFQT) in 1980. A new AFQT, in which the Math Knowledge (MK) subtest replaces NO, was recommended by the Joint Service Selection and Classification Working Group in 1986. The purpose of this research memorandum is to present percentile score norms for the proposed new AFQT in the 1980 Youth Population and the scaling of the current forms of the AFQT to the 1980 score scale.

Hands-on tests of job performance have intrinsic validity because of their high fidelity to job behavior. However, they are susceptible to poor content and measurement validity. The purpose of this analysis is to examine the content and measurement validity of prototype hands-on tests developed for three Marine Corps specialties - Ground Radio Repair, Automotive Mechanic, and Infantry Rifleman.