ERIC Number: ED211594
Record Type: Non-Journal
Publication Date: 1980
Reference Count: N/A
An Introduction to Rasch's Measurement Model.
Some basic concepts of the one-parameter logistic latent-trait model, or the Rasch model, are presented. This model assumes that the probability of a correct answer to an item is a function of two parameters, one representing the difficulty of the item and one representing the ability of the subject. The purpose of this paper is to explain a mathematical-statistical solution to the problem of separating the factor of item difficulty from person ability. Rasch's "theory of specific objectivity" is basically a theory of unidimensionality. It states that only one dimension should be measured at a time and all items should be homogeneous, or measure the same ability. Local stochastic independence is assumed. The Rasch model is compared to other test-theoretic models including classical test-theory and other item-response models. In relation to a concrete example, it is demonstrated how the parameters in the model can be estimated and how the assumptions of the model can be tested. Possible areas of application of the Rasch model are discussed. (Author/DWH)
Descriptors: Academic Ability, Academic Achievement, Difficulty Level, Latent Trait Theory, Mathematical Models, Measurement Techniques, Test Items
ERIC Clearinghouse on Tests, Measurement, and Evaluation, Educational Testing Service, Princeton, NJ 08541 ($5.50).
Publication Type: ERIC Publications; Speeches/Meeting Papers
Education Level: N/A
Sponsor: Swedish Council for Research in the Humanities and Social Sciences, Stockholm.; National Inst. of Education (ED), Washington, DC.; National Swedish Board of Education, Stockholm.
Authoring Institution: ERIC Clearinghouse on Tests, Measurement, and Evaluation, Princeton, NJ.