Tested parameter estimation method suitable for self-adaptive test of hierarchical classification design

Abstract

The invention discloses a tested parameter estimation method suitable for self-adaptive testing of hierarchical classification design. The method comprises the steps that 1, the precision of estimation requirements is set; 2, randomly extracting quantitative questions from a question bank to be answered by the testee; 3, determining characteristic values C, N and M and an estimated value X of theextracted questions; 4, after initial calculation is completed, obtaining an N value range; 5, obtaining the estimated value of X again; step 6, on the premise of conforming to the N value range, correcting the estimated values of the current N value range and the tested X value again, and then extracting questions again; and step 7, after the current N value is stable, stopping the test to obtainquestion parameters of hierarchical classification of the tested questions. The invention relates to the technical field of test question parameter analysis. According to the subject parameter estimation method for the self-adaptive test suitable for the hierarchical classification design, the problem that the ability of a tester is difficult to estimate through a short test length when an existing method faces numerous hierarchical classification test data is solved.

Description

technical field [0001] The invention relates to the technical field of parameter analysis of test items, in particular to a parameter estimation method for self-adaptive tests applicable to hierarchical classification design. Background technique [0002] For self-adaptive tests, especially for hierarchical classification items, how to use item parameters to extract items suitable for the testers and estimate the parameters of the subjects is an important task. Hierarchical classification questions in self-adaptive testing need to use topic parameter extraction to adapt to the tester's questions. Hierarchical classification extraction is required, so that the parameters of testers at different levels and categories can be accurately obtained, and the length of the test should be as short as possible. The accuracy of the tester's ability estimation is guaranteed under the premise. Contents of the invention [0003] (1) Solved technical problems [0004] Aiming at the defi...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the prior art, the present invention provides a method for estimating the parameters of the subjects in the self-adaptive test applicable to the hierarchical classification design, which solves the problem that the existing method is difficult to pass through short The test length predicts the ability of the tester, and the accuracy is not high