Reviews methods for handling missing data in a research study. Model-based methods, such as maximum likelihood using the EM algorithm and multiple imputation, hold more promise than ad hoc methods. Although model-based methods require more specialized computer programs and assumptions about the nature of missing data, these methods are appropriate…

L. A. Goodman's loglinear formulation for bi-way contingency tables is extended to tables with or without missing cells and is used for exploratory purposes. Three-way tables and generalizations of correspondence analysis are deduced, and a generalized version of Goodman's algorithm is used to estimate scores in all cases. (Author/TJH)

This paper is concerned with estimation and hypothesis testing of treatment effects in nonequivalent control group designs with the assumption that in the absence of treatment effects, natural growth conforms to a particular class of continuous growth models. Point estimation, interval estimation, and hypothesis testing procedures were developed…

Selectivity problems are discussed in terms of a general model that is estimated by the maximum likelihood method. Both single-group and multiple-group analyses are considered. An extension of the general model to latent variable models is discussed. (Author/PN)

A class of multidimensional latent trait models is described. The properties of the model parameters, and initial results on the accuracy of a maximum likelihood procedure for estimating the model parameters are discussed. The model presented is a special case of the general model described by Rasch (1961), with close similarities to the models…

The dynamic linear model or Kalman filtering model provides a useful methodology for predicting the past, present, and future states of a dynamic system, such as an object in motion or an economic or social indicator that is changing systematically with time. Recursive likelihood methods for adaptive Kalman filtering and smoothing are developed.…

A crossed random effects model is presented that applies to data with a nested structure to provide maximum likelihood estimates through the EM algorithm. The procedure is illustrated in studies of neighborhood and school effects on educational attainment in Scotland and classroom effects on mathematics learning in the United States. (SLD)

Often, all of the variables in a model are latent, random, or subject to measurement error, or there is not an obvious dependent variable. When any of these conditions exist, an appropriate method for estimating the linear relationships among the variables is Least Principal Components Analysis. Least Principal Components are robust, consistent,…

A strategy is described for dealing with measurement situations in which certain categories of responses are null, that is, persons do not respond in certain categories to certain items. The method is described for the partial credit model while maintaining the integrity of the original response framework. (SLD)

This study examines the application of the marginal maximum likelihood (MML) EM algorithm to the parameter estimation problem of the three-parameter normal ogive and logistic polychotomous item response models. A three-parameter normal ogive model, the Graded Response model, has been developed on the basis of Samejima's two-parameter graded…

Three strategies for augmenting the interpretation of significance test results are illustrated. Determining the most suitable indices to use in evaluating empirical results is a matter of considerable debate among researchers. Researchers increasingly recognize that significance tests are very limited in their potential to inform the…

Work on item response theory was extended to two areas not extensively researched previously, including models for: (1) test items that require more than one ability for a correct response (MIRT); and (2) interaction between modules of instruction that have a hierarchical relationship (HST). In order to develop the MIRT and HST models, the author…

Methods for examining the viability of assumptions underlying generalized linear models are considered. By appealing to the likelihood, a natural generalization of the raw residual plot for normal theory models is derived and is applied to investigating potential misspecification of the linear predictor. A smooth version of the plot is also…

A solution to the problem of specification error due to excluded variables in statistical models of treatment effects in nonrandomized (nonequivalent) control group designs is presented. It involves longitudinal observation with at least two pretests. A maximum likelihood estimation program such as LISREL may provide reasonable estimates of…

The development of a probabilistic model for validating Gange's learning hierarchies is described. Learning hierarchies are defined as paired networks of intellectual tasks arranged so that a substantial amount of positive transfer occurs from tasks in a lower position to connected ones in a higher position. This probabilistic validation technique…