Research for Personnel Classification Tests

In times of budgetary constraint, the Marine Corps must assess its need for high aptitude enlisted recruits. Ideally, hands-on performance tests would provide the benchmark measure of job performance against which the Marine Corps could assess its enlistment aptitude standards and the proficiency of its personnel. However, hands-on performance tests are too expensive to use for ongoing monitoring of aptitude standards and evaluation of training needs. Surrogate measures, such as job knowledge test, might provide performance benchmarks at a fraction of the cost of hands-on tests. This paper assesses the usefulness of several surrogate job performance measures for evaluating enlistment aptitude standard and determining USMC training needs.

Distributions of test scores need to be smoothed in equating and/or norming. Popular parametric smoothing procedures are based on beta-binomial and log-linear models. A new approach has been developed using polynomials of the beta-binomial cumulative distribution function. The same approach was also applied to extend the beta-binomial family to more than four parameters. These methods were compared using cross-validation in two examinee samples who took the Armed Services Vocational Aptitude Battery. Results show that the log-linear and extended beta-binomial families fit the data about equally well.

Two forms, each containing 35 verbal and 30 mathematics items, have been developed for a new Enlistment Screening Test (EST) to predict Armed Forces Qualification Test (AFQT) scores of military applicants. These forms were constructed in two stages from items in discontinued versions of the Defense Department's test batteries. The first stage was to develop overlength forms from the available item pool. This research memorandum describes the second stage: constructing final forms by selecting items from the overlength forms. Item selection was based on the correlation of the item with AFQT, in a subsample of applicants with AFQT percentiles between 21 and 65. For each EST form, the AFQT score was predicted from the total score on the final EST items. The results were used to calculate expectancy tables which, for any given EST score, provide probabilities of exceeding the specified AFQT cutoffs. These probabilities are reported in tables.

Overlength forms, containing about 50 percent more items than needed in the final forms, have been developed for a new Enlistment Screening Test. These forms were constructed using items from discontinued forms of the DoD's test batteries. This research memorandum describes the data analyses and their results.