Statistical Tests for Conditional Independence in a Hierarchial Model for Speed and Accuracy on Test Items |
Contents
Executive Summary | 1 |
Generalization to HigherOrder Independencies | 8 |
FirstOrder and SecondOrder Derivatives | 14 |
Copyright | |
Common terms and phrases
3PL model 96 items 96 Significance Probability a²l additional parameter adjacent pairs alternative model assumptions in 11 assumptions of conditional bivariate normal distribution computerized adaptive testing conditional independence covariance matrix degrees of freedom derivatives diagonal diagonal matrix Distribution of outcomes Distribution of significance elements a² estimation evaluated first-level models Frequency Frequency H₁ heuristic argument hierarchical framework Hierarchical Model independence assumption Independence between responses individual item IRT model item parameters Lagrange multiplier Linden LM Pik LM statistic LM tests LM(k loglikelihood lognormal model maximum likelihood model for speed multivariate normal null hypothesis pair of items parameter vector parameterization person parameters Pikj respectively response and RT response model responses given RT distributions sect Significance Probability LM single test taker speed and accuracy Statistical Tests subset of items t₁ Test for Independence test items test taker's ability test the hypothesis type of conditional U₁ VijVkj violations στ