Open AccessCOMPUTERIZED MASTERY TESTING USING FUZZY SET DECISION THEORY
Author(s) -
Du Yi,
Lewis Charles,
Pashley Peter J.
Publication year - 1994
Publication title -
ets research report series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.235
H-Index - 5
ISSN - 2330-8516
DOI - 10.1002/j.2333-8504.1994.tb01610.x
Subject(s) - rasch model , fuzzy set , fuzzy logic , bayesian probability , set (abstract data type) , item response theory , computer science , artificial intelligence , decision rule , bayesian statistics , statistical hypothesis testing , test (biology) , machine learning , mathematics , bayesian inference , statistics , psychometrics , paleontology , biology , programming language
Lewis and Sheehan (1990) developed a computerized sequential mastery test procedure that utilizes Bayesian decision theory to make “master,” “nonmaster,” or “continue testing” decisions. While maintaining their general framework of administering sequential testlets, a fuzzy set approach was used to develop an alternative computerized mastery test. This new procedure differs from Lewis and Sheehan's in that the decision rule is determined using fuzzy set decision theory, and ability estimates are obtained using the Rasch model rather than a three‐parameter logistic model. This article describes this new approach and illustrates the differences between the fuzzy set and Bayesian methods by way of an example.