Random forests are a nonparametric machine learning method, which is currently gaining popularity in the behavioral sciences. Despite random forests' potential advantages over more conventional statistical methods, a remaining question is how reliably informative predictor variables can be identified by means of random forests. The present study…

We extend a unified and easy-to-use approach to measurement error and missing data. In our companion article, Blackwell, Honaker, and King give an intuitive overview of the new technique, along with practical suggestions and empirical applications. Here, we offer more precise technical details, more sophisticated measurement error model…

We study the properties of two specification tests that have been applied to a variety of estimators in the context of value-added measures (VAMs) of teacher and school quality: the Hausman test for choosing between random and fixed effects and a test for feedback (sometimes called a "falsification test"). We discuss theoretical…

A difficult result to interpret in Computerized Adaptive Tests (CATs) occurs when an ability estimate initially drops and then ascends continuously until the test ends, suggesting that the true ability may be higher than implied by the final estimate. This study explains why this asymmetry occurs and shows that early mistakes by high-ability…

This report describes work using nationally representative 2005 data from the School Nutrition Dietary Assessment-III (SNDA-III) study to develop a simulation model to predict the potential implications of changes in policies or practices related to school meals and school food environments. The model focuses on three domains of outcomes: (1) the…

Developed a simulation model to predict the effects of policies for reducing adolescent smoking initiation. The model projected numbers of smokers, never smokers, and ex-smokers by age, sex, and race/ethnicity and effects of reductions in youth initiation. Results predicted that even if tobacco policies eliminated youth initiation, the number of…

The purpose of this study was to develop a screening procedure for undergraduate pilot training (UPT). This procedure was based upon the use of ground-based instrument trainers in which UPT candidates, naive to flying, were evaluated in their performance of job sample tasks; i.e., basic instrument flying. Training and testing sessions were…

Used a mathematical simulation procedure adaptable to an interactional concept of personality to predict the interpersonal compatibility of couples. Strife scores derived from computer simulation of interactional personality data correlated significantly with partner ratings for the quality and the stability of their relationship. Significance…

Predictors of novice learning in simulation environments were investigated in the domain of statistics. The first objective was to clarify the relation between intellectual ability and working method (e.g., orientation and systematical orderliness), and to determine the effects on learning of working method, independently of intellectual ability.…

Several advantages to the use of factor scores as independent variables in a multiple regression equation were found. To help select the most desirable type of factor score on which to calculate a regression equation, computer-based Monte Carlo methods were used to compare the predictive accuracy upon replication of regression of five…

Using simulated item response data, the performance of several "robust" and conventional schemes for ability estimation was evaluated in conjunction with logistic item response theory models (one, two, and three parameter models). The simulated item response data were generated using a model that is more complex than are the usual…

This paper reviews the existing multidimensional item response theory (IRT) models and demonstrates how one of the models can be applied to estimation of abilities from a test measuring more than one dimension. The purposes of this paper were threefold. First, the fundamental concepts required when considering multidimensional models for the…

An article was published in the December 1985 issue of this journal that used LISREL to investigate the relationship between family medicine residents' moral reasoning and their clinical performance (Sheehan). The data from that study are reanalyzed to illustrate the importance of considering alternative structural equation models in evaluation…

Two studies that compare the use of analysis of patterns in time (APT) (formerly non-metric temporal path analysis) and the linear models approach (LMA) are reported. In the first study, scores on the College Entrance Examination Board French Achievement Test (CEEBFAT), number of years of study of high school French (YHSF), and course achievement…

A multiparameter, programmable model was developed to examine the interactive influence of certain parameters on the probability of deciding that an examinee had attained a specified degree of mastery. It was applied within the simulated context of performance testing of military trainees. These parameters included: (1) the number of assumed…