The Comparison And Application Of Computerized Multistage Adaptive Routing Rules

Computerized multistage adaptive testing(MST) is a combination of conventional paper-and-pencil(P&P) and item level computerized adaptive test(CAT), it is a kind of test forms based on computerized technology, take the item set as the scoring unit. It can be said that, MST estimate the Ability extreme value more accurate than conventional paper-and-pencil(P&P),and also used the CAT auto-adapted characteristic to reduce the examination length and the score time of report.Routing methods is the key issues of MST, it directly affect the effectiveness of its management.Based on the accurate measurement of the examinee ability, research using the monte carlo simulation to compare the 1-3-3, 1-3-5, 1-5-5 experiment under the condition of three kinds of routing rules based on MSE(NC, Proximity, TI), and estimate the accuracy, the correct classification ratio and module exposure of three routing methods.Today, intelligence test is high degree in society, and it is developed. In the field of education, intelligence test is taken seriously too. Therefore, in education activity, teachers should take different method for teaching different students, and intelligence test is to measure its characteristics, one of the effective ways to quantify student’s characteristics. Then, developing and improving the intelligence test, is very necessary.In addition, with the improvement of modernization in our country, the application of computer in teaching has become widespread.Using the advantage of Raven’s Standard Progressive Matrices(RSPM)about without being limited by the culture and language characteristics, this research applicate the RSPM into the 1-3-3-3 Computer multistage adaptive testing.In order to optimize the item, reduce the test time and fit the need of modern teaching.After the comparison of three routing methods and application the MST, we made three conclusions:1、In terms of capacity estimation accuracy and robustness, TI is the best, NC is the worst. On the uniformity of module Proximity is the same as TI, NC is the worst.2、In the limited-time MST of simulated empirical study, TI is the best in capacity estimation accuracy and robustness, NC is the worst. Among them, the Proximity is as the same as TI.3、In the limited-time MST RSPM of empirical study, capacity distribution is in normal. In addition, the validity of limited-time MST is high, so it can be used as an intelligence tool.