【摘 要】
：
A testlet is a cluster of items that share a common stimulus,and the possible local dependence among items within a testlet is called testlet-effect.Under the framework of Item Response Theory(IRT),va
【机 构】
：
Beijing Normal University,China The Hong Kong Ins
论文部分内容阅读
A testlet is a cluster of items that share a common stimulus,and the possible local dependence among items within a testlet is called testlet-effect.Under the framework of Item Response Theory(IRT),various Testlet Response Models(TRM) have been developed to take into account such testlet-effect(e.g.,Bradlow,Wainer,& Wang,1999;Wang & Wilson,2005;Li,Blot,& Fu,2006).
其他文献
Self-esteem is a widely studied construct in psychology that is typically measured by the Rosenberg Self-Esteem Scale(RSES).
Mixture IRT modeling is a promising approach,which is a combination of Item Response Theory(IRT) and Latent Class Analysis(LCA).It can overcome disadvantages of IRT and LCA,but also keep advantages of
Bifactor models have a long history yet have been controversial(Reise,2012).In bifactor models,each indicator loads on a general factor and one of the group factors,and thus the main effect and second
An ideal item pool consists of an appropriate number of item combinations that meet all content specifications of a test and provide sufficient estimation for examinees at different proficiency levels
Calibration of new items online has been an important topic in both psychological and educational measurement.Recently,four online calibration methods were developed in the context of Multidimensional
Online calibration is a technology-enhanced strategy for pretesting new items in Computerized Adaptive Tests(CATs).Many CATs are administered continuously over a long-term and rely on large item banks
Three-way principal component analysis(3WPCA) techniques for a data array of objects by variables by sources can be formulated with the core matrices that explain the inter-sources differences in the
Principal Cluster Components Analysis(PCCA) is a procedure to cluster variables to define linear composites with maximal explained variances based on the matrix of correlations between variables.
Item Response Theory(IRT) is a widely used statistical modeling method in psychometrics,primarily to estimate item and person parameters to render the true ability(or endorsement) interpretable.
There are some studies in the literature that have considered the impact of modeling multidimensional mixed structured tests as unidimensional.These studies have determined that the error associated w