一种多策略认知诊断方法:MSCD方法的开发

A New Multiple-Strategies Cognitive Diagnosis Model:the MSCD Method

当前国内外开发的认知诊断模型基本上只能处理单策略的测验情景,并假设所有被试均采用同一种加工策略/解题策略,从而忽视了加工策略的多样性及差异性。本研究根据de la Torre和Douglas(2008)采用多个Q矩阵来表征多个加工策略的思想,并结合使用丁树良等(2009)修正的Q矩阵理论及孙佳楠,张淑梅、辛涛和包珏(2011)的广义距离判别法,开发了一种新的多策略认知诊断方法——MSCD方法。Monte Carlo模拟研究结果表明:在单策略测验情景下,传统的单策略认知诊断方法与采用MSCD方法的诊断正确率均比较理想,且差异不大;但在多策略测验情景时,传统的单策略认知诊断方法诊断正确率较低,而MSCD方法的诊断正确率却仍较理想;当加工策略增至5种时,MSCD方法仍有较高的边际判准率、模式判准率以及加工策略判准率。研究表明MSCD方法基本合理、可行。这为实现对加工策略的诊断提供了方法学支持,有利于拓展认知诊断在实际中的应用。

Almost all of the current cognitive diagnosis models allow for single-strategy of problem solving, and assume all examinees use the same processing strategy. Based on the extant studies, the current study developed a new multiple-strategies cognitive diagnosis model, called MSCD method. Monte Carlo method was employed here to explore the feasibility of the MSCD method and to examine the estimation precision as well as the properties of the MSCD method. The findings were presented： （1） Under the single-strategy tests or data, the average attribute match ration （AAMR）, pattern match ration （PMR） and strategy match ration （SMR） of the MSCD method were as high as those of single-strategy cognitive diagnosis model （single-strategy model）. （2） Under multiple-strategy tests or data, the MSCD method also had high AAMR, PMR and SMR, but single-strategy model did not. （3） When the number of strategy reaching 5, the MSCD method also had very high AAMR, PMR and SMR. All in all, the new developing model, the MSCD method, was reasonable and could be accepted.