Classical test theory, item response theory, and generalizability theory all treat the abilities to be measured as continuous variables, and the items of a test as independent probes of underlying continua. These models are well-suited to measuring the broad, diffuse traits of traditional differential psychology, but not for measuring the outcomes…

It has been demonstrated that the individual variation in the level and rate of learning for a cohort of students over time can be estimated by hierarchical linear models. Models of this type can also be estimated using widely available structural modeling software, which provides a flexible framework for model explorations, including the use of…

This study investigates the amount of uncertainty added to National Assessment of Educational Progress (NAEP) estimates by equating error under both ideal and less than ideal circumstances. Data from past administrations are used to guide simulations of various equating designs and error due to equating is estimated empirically. The design…

Tables of specifications are used to guide test developers in sampling items and maintaining consistency from form to form. This paper is a generalizability study of the American College Testing Program (ACT) Achievement Program Mathematics Test (AAP), with the content areas of the table of specifications representing multiple dependent variables.…

The inherent methodological problem in conducting research at multiple sites is how to best derive an overall estimate of program impact across multiple sites, best being the estimate that minimizes the mean square error, that is, the square of the difference between the observed and true values. An empirical example illustrates the use of the…

Advantages and disadvantages of standard Rasch analysis computer programs are discussed. The unconditional maximum likelihood algorithm allows all observations to participate equally in determining the measures and calibrations to be obtained quickly from a data set. On the advantage side, standard Rasch programs can be used immediately, are…

Standards, passing scores, or cut scores have been seen as an element of criterion-referenced tests since their introduction. This paper discusses at least two issues surrounding the establishment of cut scores which appear to need clarification: (1) the theoretical definition of a cut score; and (2) decisions which must be made in selecting a…

This study is an attempt at a cohesive characterization of the concept of essay reliability. As such, it takes as a basic premise that previous and current practices in reporting reliability estimates for essay tests have certain shortcomings. The study provides an analysis of these shortcomings--partly to encourage a fuller understanding of the…

A primary goal of this study was to determine the extent to which item difficulty was related to item position and, if a significant relationship was found, to suggest adjustments to predicted item difficulty that reflect differences in item position. Item response data from the Medical College Admission Test (MCAT) were analyzed. A data set was…

A scoring rubric transition study is described. It was designed to evaluate possible drift in scoring the Maryland Writing Test from year to year (when using a modified holistic scoring method), to evaluate strategies for revising swing rubrics from narrative and explanatory writing while maintaining original scoring standards, and to establish…

If a loss function is available specifying the social cost of an error of measurement in the score on a unidimensional test, an asymptotic method, based on item response theory, is developed for optimal test design for a specified target population of examinees. Since in the real world such loss functions are not available, it is more useful to…

The sampling errors of maximum likelihood estimates of item-response theory parameters are studied in the case where both people and item parameters are estimated simultaneously. A check on the validity of the standard error formulas is carried out. The effect of varying sample size, test length, and the shape of the ability distribution is…

The purpose of this study was to evaluate the utility of out-of-level testing (OLT) when it is applied to the assessment of special education students with mild learning handicaps. This evaluation of OLT involved testing hypotheses related to: (1) the adequacy of vertical scaling, (2) the reliability and (3) the validity of OLT scores. Fifty-eight…

Covariance structure modelling is applied to the problem of estimating reliability and measurement error in survey data. To provide a basis for grouping certain question or variable types (data from questions), a simple typology based on the formal characteristics of the questions is outlined. From this classification, models for the different…

How accurately ability is estimated when the test model does not fit the data is considered. To address this question, this study investigated the accuracy of the maximum likelihood estimator of ability for the one-, two- and three-parameter logistic (PL) models. The models were fitted into generated item characteristic curves derived from the…