The potential of item response theory (IRT) for solving a number of testing problems in the Maryland Functional Reading Program would appear to be substantial in view of the many other promising applications of the theory. But, it is well-known that the advantages derived from an IRT model cannot be achieved when the fit between an item response model and the test data of interest is less than adequate. The principal purpose of the research reported in this paper was to investigate the fit of the one-, two-, and three-parameter logistic models to the test results obtained from the administration of the 1982 Maryland Functional Reading Test (MFRT). The evidence addressing model-data fit seemed clear: a two-parameter logistic model was able to adequately account for examinee performance on the MFRT. The one-parameter model could not handle the substantial variation among test items in their discriminating power. The three-parameter model improved the fit only slightly because of the minimum amount of guessing on the test. Several suggestions were offered in the paper for conducting goodness-of-fit investigations. (Author)