DESCRIBING AND CATEGORIZING DIF IN POLYTOMOUS ITEMS
نویسندگان
چکیده
منابع مشابه
Feature Pattern Analysis for Polytomous Items
Feature Pattern Analysis (FPA), as a part of the theory of data (Coombs, 1964), can be interpreted as a formal model of some procedures frequently applied in the behavioral and social sciences, e.g., for Guttman scaling, Parallelogram Analysis, Likert scaling, the Semantic Differential, and the Kelly Grid. FPA may also be used as a scaling procedure for nominal and ordinal data with qualitative...
متن کاملConditional Covariance Theory and DETECT for Polytomous Items
This paper extends the theory of conditional covariances to polytomous items. It has been mathematically proven that under some mild conditions, commonly assumed in the analysis of response data, the conditional covariance of two items, dichotomously or polytomously scored, is positive if the two items are dimensionally homogeneous and negative otherwise. The theory provides a theoretical found...
متن کاملA Multinomial Error Model for Tests with Polytomous Items∗
This paper introduces a multinomial error model, which models the raw scores from polytomously scored items. The multinomial error model is implemented in this paper for estimating conditional standard errors of measurement and reliability for both raw and scale scores. A simulation study is presented, which suggests that the multinomial conditional standard errors of measurement for the raw an...
متن کاملA Strong True-Score Model for Polytomous Items∗
This paper presents a strong true-score model for tests that consist of polytomously-scored items. Errors conditional on true score are assumed to be distributed as multinomial, true category-proportion scores are assumed to follow a Dirichlet distribution, and the marginal category scores are the Dirichletmultinomial distribution. The model is illustrated using real data sets in the following ...
متن کاملEstimation of Models in a Rasch Family for Polytomous Items and Multiple Latent Variables
The Rasch family of models considered in this paper includes models for polytomous items and multiple correlated latent traits, as well as for dichotomous items and a single latent variable. An R package is described that computes estimates of parameters and robust standard errors of a class of log-linear-by-linear association (LLLA) models, which are derived from a Rasch family of models. The ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: ETS Research Report Series
سال: 1997
ISSN: 2330-8516
DOI: 10.1002/j.2333-8504.1997.tb01726.x