Educational measurement-oriented objective test question attribute mode estimation and correction method and educational measurement-oriented objective test question attribute mode estimation and correction system

Abstract

The invention belongs to the technical field of education measurement and education data mining, and discloses an objective test question attribute mode estimation and correction method and system foreducation measurement. The method comprises the steps: training a measurement matrix M1 for supervised measurement learning, and employing the matrix to preliminarily mine the knowledge mastering conditions of all learners; taking part of the data as a training set, training a measurement matrix M2 for supervised measurement learning, estimating an attribute mode of a new objective test question,and then adding the test question as a known test question into the basic objective test question until the attribute modes of all test questions are estimated; after the test questions are estimated, one test question is selected for attribute mode correction, and the corrected objective test question attribute mode is used for mining knowledge mastering conditions, namely learner attribute modes, of all learners. Teaching personnel are helped to analyze test question composition more objectively and detailedly, the test validity and reliability are improved, and a basis is provided for learners to adjust learning plans or take targeted remedial measures.

Description

technical field [0001] The invention belongs to the technical field of educational measurement and educational data mining, and in particular relates to a method and system for estimating and correcting objective test question attribute patterns oriented to educational measurement. Background technique [0002] At present, the closest existing technology: With the rapid development of cognitive psychology, psychometrics and computer technology, cognitive diagnosis theory can make accurate assessments of learners in the field of education and psychometrics from the perspective of micro cognition. Evaluation and feedback, this advantage shows the great development potential of cognitive diagnostic theory. Among them, the description of the relationship between test items and cognitive attributes is called the test Q matrix, which is the theoretical basis for realizing the advantages of accurate assessment and feedback for learners from the perspective of micro-cognition. Howe...

AI Technical Summary

Problems solved by technology

The purpose of the present invention is to solve the problem that in the prior art, relying on experts to mark the Q matrix has greater subjectivity, and the workload of manual marking is large, and the accuracy is poor; Based on the premise of model-data fitting, the defined Q matrix has a large error, which in turn affects the subsequent parameter mining of test items. The mining of learner attribute patterns is inaccurate, and it is difficult to be used in other models. The applicability is poor; When calculating the pattern similarity and learner attribute pattern similarity, for the selection of objective test item attribute pattern or learner attribute pattern that leads to the same minimum distance, the method of random selection is adopted, and there is no reasonable basis for judgment, so it is impossible to make such a classification interpretation, and can lead to misclassification of patterns

Images