The Armed Services Vocational Aptitude Battery (ASVAB), a battery of 10 vocational tests, was administered to a national probability sample of nearly 12,000 young people, 15 to 23 years of age. Score distributions from the study were needed by the Armed Services to set performance levels for accepting and assigning recruits. Three psychometric models designed to provide information about the operating characteristics of the tests and the items of which they are comprised were fit to the eight "power" tests of the ASVAB, based on a random sample of 1,187 respondents from the national study. The models in question were the one-, two-, and three-parameter item response models that were fit to the data by the marginal estimation approach by means of the BILOG computer program. Results for each model and test included indices of overall goodness-of-fit, item parameter estimates and associated standard errors, item information indices, and test information and standard error curves. Appendixes contain the results of the analyses. Appendix A contains classical item statistics for each of the tests, including item percents-correct and item-test correlation, both Pearson and biserial. Appendixes B through J present results from the BILOG runs under the one-, two-, and three-parameter models. (YLB)