2. 您的位置
3. 首页
4. 会议论文
5. Evaluating t-distribution and sandwich robust methods for level-1 outliers in hierarchical linear mo

Evaluating t-distribution and sandwich robust methods for level-1 outliers in hierarchical linear mo

来源 :北京师范大学 | 被引量 : 0次 | 上传用户：QCLHQCLH

【摘 要】

：

The Hierarchical Linear Model(HLM) has become popular and widely used in various educational studies recently.However,violations of normality assumptions can have a non-ignorable impact on the model e

【作 者】

：

Zhenqiu (Laura) Lu Jue Wang 

【机 构】

：

University of Georgia,USA

【出 处】

：

北京师范大学

【发表日期】

：

2015年5期

下载到本地 , 更方便阅读

下载此文 赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

The Hierarchical Linear Model(HLM) has become popular and widely used in various educational studies recently.However,violations of normality assumptions can have a non-ignorable impact on the model estimation.Both point estimates and standard errors for fixed and random effects can be sensitive to outliers at all levels.

其他文献

Bayesian structural equation modeling approach to confirming testlet effects:Is a bifactor structure

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

会议

Enhancing item pool utilization for designing multistage computerized adaptive tests

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

会议

A comparative study of online item calibration methods in multidimensional computerized adaptive tes

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

会议

Exploring online calibration of polytomous items in computerized adaptive testing

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

会议

Sparse core Tucker2 for computationally identifying the optimal model between parafac and Tucker2

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

会议

Applications of principal cluster components analysis to scale development and the validation of clu

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.

会议

Component-based item response theory

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.

会议

The interaction effect of the correlation between dimensions and item discrimination on parameter es

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

会议

The logistic testlet framework for within-item multidimensional testlet-effect

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

会议

Estimating multi-level models in data streams

Multi-level models have found numerous applications in the social sciences for the analysis of grouped data.Mostly,these models are fit to a static dataset.Recent technological advances in the measure

会议