The cross-classified multilevel measurement model: an explanation and demonstration

The link between the hierarchical generalized linear model (HGLM) and the Rasch model's parameterization has already been demonstrated by several researchers. Extensions have been described that include higher clustering levels to model more appropriately the contextual effects that are frequently encountered in educational research. However, pure hierarchies are relatively rare and instead cross-classified data structures are more frequently encountered. Cross-classified random effect modeling (CCREM) is still not commonly used. Use of CCREM in combination with the multilevel measurement model (MMM) has been recently introduced and is described further in the current study. Specifically, the link between the MMM and the CCREM MMM (termed "CCMMM" model) is provided. A dataset was simulated to demonstrate interpretation of the CCMMM model's parameters and to compare results under a CCMMM versus HGLM analysis. An Appendix is provided to demonstrate SAS GLIMMIX code used to estimate HGLM and CCMMM models' parameters.     

The CDF Level 3 trigger

We discuss the architecture, operation, and performance of the Level 3 trigger used in the Fermilab CDF experiment in 1988–1989. This trigger used an on-line computer farm of 58 Motorola 68020 processors operating in parallel. The on-line programs were written mainly in Fortran and run in an environment similar to that used for off-line analysis.     

Moving the cut score on Rasch scored tests

Empirically based item selection guidelines are presented for moving the cut score on equated tests consisting of n dichotomous items calibrated assuming the Rasch model. The cut score on a test form B, c(B), may be made higher than test form A's cut score, c(A), in the following ways: (1) select items for test form B such that the variance of test form B's item difficulties, sigma(2)(B), will be equal to test form A's sigma(2)(A), but test form B's mean item difficulty, mu(B), will be less that of test form A, mu(A); (2) given c(A) > n/2, select items for test form B such that mu(B) s(2)(A). To make c(B) lower than c(A), the direction of the changes listed above for the two tests item difficulties sigma(2) and mu should be reversed. Derivations of lemmas that underlie the guidelines are provided as well as a simulated example.     

The multilevel measurement model: introduction to the special issue

An introduction to the special issue on the multilevel measurement model (MMM) is provided. The two- and three-level multilevel models for continuous outcomes are reviewed. The extension to the hierarchical generalized linear model and its use as a multilevel measurement model for dichotomous measurement indicators is demonstrated. The six articles in the special issue are described.     

Methodological challenges encountered in summarizing evidence-based practice.

This paper details the challenges encountered by authors summarizing evidence from a primary study to describe a treatment's effectiveness using an effect size (ES) estimate. Dilemmas that are encountered, including how to calculate and interpret the pertinent standardized mean difference ES for results from studies of various research designs, are described. Recommendations are offered to authors of primary studies and to those conducting summaries of primary studies. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Meta-Analytic Methods of Pooling Correlation Matrices for Structural Equation Modeling Under Different Patterns of Missing Data.

Three methods of synthesizing correlations for meta-analytic structural equation modeling (SEM) under different degrees and mechanisms of missingness were compared for the estimation of correlation and SEM parameters and goodness-of-fit indices by using Monte Carlo simulation techniques. A revised generalized least squares (GLS) method for synthesizing correlations, weighted-covariance GLS (W-COV GLS), was compared with univariate weighting with untransformed correlations (univariate r) and univariate weighting with Fisher's z-transformed correlations (univariate z). These 3 methods were crossed with listwise and pairwise deletion. Univariate z and W-COV GLS performed similarly, with W-COV GLS providing slightly better estimation of parameters and more correct model rejection rates. Missing not at random data produced high levels of relative bias in correlation and model parameter estimates and higher incorrect SEM model rejection rates. Pairwise deletion resulted in inflated standard errors for all synthesis methods and higher incorrect rejection rates for the SEM model with univariate weighting procedures. (PsycINFO Database Record (c) 2010 APA, all rights reserved)