This Article Full Text (OnlineFirst PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Scopus Google Scholar Articles by Almond, R. G. Articles by Yan, D. Social Bookmarking                         What's this?

Bayesian Network Models for Local Dependence Among Observable Outcome Variables

^* To whom correspondence should be addressed. E-mail: ralmond{at}ETS.ORG.

	Abstract

Bayesian network models offer a large degree of flexibility for modeling dependence among observables (item outcome variables) from the same task, which may be dependent. This article explores four design patterns for modeling locally dependent observations: (a) no context—ignores dependence among observables; (b) compensatory context—introduces a latent variable, context, to model task-specific knowledge and use a compensatory model to combine this with the relevant proficiencies; (c) inhibitor context—introduces a latent variable, context, to model task-specific knowledge and use an inhibitor (threshold) model to combine this with the relevant proficiencies; (d) compensatory cascading—models each observable as dependent on the previous one in sequence. This article explores the four design patterns through experiments with simulated and real data. When the proficiency variable is categorical, a simple Mantel-Haenszel procedure can test for local dependence. Although local dependence can cause problems in the calibration, if the models based on these design patterns are successfully calibrated to data, all the design patterns appear to provide very similar inferences about the students. Based on these experiments, the simpler no context design pattern appears more stable than the compensatory context model, while not significantly affecting the classification accuracy of the assessment. The cascading design pattern seems to pick up on dependencies missed by other models and should be explored with further research.

First published on October 26, 2009
Journal of Educational and Behavioral Statistics 2009, doi:10.3102/1076998609332751


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?