认知诊断是认知心理学与心理计量学相结合的产物，它有助于人们更好地了解人类内部心理活动规律及加工机制，实现对个体认知发展实况(含优点与缺陷)的诊断评估，以促进个体全面发展。本项目拟对当前具有重要理论意义和应用价值的多级计分认知诊断进行研究，采用数学推导和建模以及多学科交叉研究方式：(1) 开发可处理多级计分数据、功能强大的广义多级计分认知诊断模型(G-PCDM)，以弥补传统模型的不足；(2) 探讨G-PCDM的参数估计方法并开发出其参数估计的算法程序，供实践者使用；(3) 构建多级计分认知诊断计算机化自适应测验(CAT)相关算法及技术(含参数估计方法、选题算法及曝光控制技术等)，以充分发挥CD和CAT双重优势，从而实现对被试高效、快速、准确地诊断评估。总之，本课题拟弥补当前国内外关于CD及其CAT研究的不足，进一步提升认知诊断的效率和准确率，并促进CD和CAT更好地为实践服务提供方法学支持。

In recent years, cognitive diagnosis (CD) has been getting more attention in education evaluation, psychological measurement and other disciplines (Rupp & Templin, 2008). Compared to the traditional test, CD tries to reveal each student’s specific cognitive strengths and weaknesses rather than providing a total score or ability value. In additional, cognitive diagnostic assessment not only can be used to evaluate students’ academic level, but also can provide diagnostic information for each student, further help students have effective remedial instruction..At present, only a few polytomous CDMs have been developed to deal with polytomous items, including the general diagnostic model (GDM,von Davier, 2008), the polytomous LCDM (Hansen, 2013), and the sequential G-DINA model (Ma & de la Torre, 2016). However, these CDMs, except the sequential G-DINA model, do not consider the relationship between attributes and response categories by assuming that all response categories of an item measure the same set of attributes. This may result in loss of diagnostic information, because different response categories could measure different attributes. More importantly, the existing polytomous models are based on different theoretical assumptions, which belong to different types of models. For example, the polytomous LCDM belongs to graded response models based on the global (or cumulative) logit, the GDM is partial credit models that make use of the adjacent category logit, while the sequential G-DINA model is a special case of the sequential process model based on the continuation ratio logit. Therefore, for these existing CDMs, each model depends on restrictive theoretical assumptions and limits their use. .Cognitive diagnostic computerized adaptive testing (CD-CAT) combines the advantages of both cognitive diagnosis and CAT, which can make adaptive diagnosis for different individuals and provide more detailed diagnostic information on the knowledge competence of the examinees. Currently, large number of CD-CAT researches focus on the dichotomous data. In our knowledge, there are no researches on CD-CAT for polytomously-scored items/data. However, polytomously-scored items/data have been broadly used in a variety of tests for their advantages of providing more information about examinee and fewer polytomous items can achieve the same precision compared with dichotomous items (van der Ark, 2001). Therefore, it would be an interesting topic on CD-CAT with polytomously-scored items (called PCD-CAT). .To address all above issues on polytomous CDMs and its CD-CAT, this study/ project aims to: (1) propose a generalized polytomou CDMs (G-PCMD) for polytomously scored items with different link functions. The existing polytomus CDMs can be considered as special cases of the proposed G-PCMD. Simulation studies were carried out to investigate the feasibility of the proposed G-PCMD; (2) use MMLE/EM algorithm to estimate the parameters of G-PCMD and develop the corresponding program for applicators; (3) construct a framework of polytomous CD-CAT (PCD-CAT) with G-PCMD, including the construction of the item bank, the item selection method, the parameter estimation and the termination rule. More important, several promising item selection algorithms used in dichotomous CD-CAT are extended to PCD-CAT and two new item selection algorithms are developed for PCD-CAT; (4) finally, a set of real data about internet addiction was analyzed to illustrate the application of the proposed G-PCMD and PCD-CAT, as well as to verify their effectiveness..In this project, both the Monte Carlo simulation method and real data analysis are employed to verify the feasibility and effectiveness of new proposed models, algorithms, methods and technologies. All these are expected to provide the methods support on promoting the efficiency and classify accuracy and expanding the application fields.