Font Size: a A A

Development Summary Of IRT And Application In Education Measurement Of IRT

Posted on:2009-03-31Degree:MasterType:Thesis
Country:ChinaCandidate:Q ZhaoFull Text:PDF
GTID:2167360245454507Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
At present, there are two guide theories for exams. One is classical test theory (CCT), the other is Item Response Theory (IRT). CCT is based on the theory of the truth value theory of scoring, and concept system as reliability, validity, difficulty and discrimination is raised and developed. It is used to measure a test tool or the quality of a test. With the fast development of education measurement, modern measuring theory, with the core of item response theory, has been the main object of the study of education measurement. It has aroused more and more attention of the researchers and educators. Compared to CCT, IRT has obvious advantage. The item parameters it takes will not be influenced by samples, and the acquisition of the parameters will not change because of the change of the samples which were taken as testing items. Meanwhile, the estimate of the tested's ability will not change with different test questions. Just because of these advantages, it is widely used in psychology and education measurements.In this thesis, the author introduced the developing history of IRT. Three-parameter Logistic Model is taken as a typical model to show the basic theory of IRT model. It is compared with CCT, and the historical background, early development, the cooperation between FA and IRT, multidimensional item response theory model and its outlook are all included in this thesis. At last, the author introduced the realization of the estimate of IRT parameters through EM algorithm. Final examination scores of 200 students from three junior middle schools in Anshan helped the workout of the parameter, and the result was analyzed. The result got from the experiment could work to tell the exam items, to analyze the students studies, and will be good for teaching and examinations.
Keywords/Search Tags:Item Analysis, IRT, Multidimensional Item Response Theory, Multilevel Scoring, Item Parameter, EM Algorithm
PDF Full Text Request
Related items