认知诊断计算机自适应测验中平衡属性收敛的新方法

提出两种认知诊断计算机自适应测验下平衡属性收敛的新方法（MABI、RTA）,模拟研究系统探讨和比较了此二者与已有方法（ABI、IABI和RABI）的表现。结果发现：（1）新方法较不考虑属性收敛的方法有更高的准确率以及更均衡的题目使用率;（2）新方法较ABI和RABI有稍低的准确性,但有更平衡的题目使用率;（3）新方法与IABI的准确性和题目使用率在不同选题策略下各有合优势。总之,两种新方法较好地兼顾测量准确性、题目使用率以及题库曝光情况。     

The Two-Sense Reading of Spinoza’s definition of attribute

Spinoza’s definition of ‘attribute’ (E1d4) has been described as ‘one of the most puzzling passages in the Ethics’ and ‘a longstanding worry’ for Spinoza interpreters. Its puzzling status stems from (a) its apparent ‘subjectivist’ character and (b) the dominant understanding of Spinoza’s notion of attribute as an ‘objectivist’ notion. The paper aspires to remove this puzzlement by proposing and defending a reading of E1d4 in which it is understood to have two senses. First, I defend the objectivist character of Spinoza’s notion of attribute, and explain why E1d4 is problematic within the framework of an objectivist interpretation of the Spinozistic attribute. Second, I propose a reading of E1d4 as having two senses and explain the cause of an apparent contradiction in E1d4. Finally, I devote the largest part of the paper to a defence of the Two-Sense Reading of E1d4. More specifically, I draw on four sources for this defence: (a) the structure of E1d1 and E1d3, (b) the content and form of E1p4 and its demonstration, (c) Spinoza’s doctrine of parallelism and (d) Descartes’ treatment of attribute and its influence on Spinoza.     

基于等级反应模型的属性层级方法

给出基于等级反应模型的属性层级方法（Attribute Hierarchy Method, AHM）,并简记为GRM-AHM,提出了相应的确定GRM-AHM的期望项目反应模式全集的方法和一种新的归类法LL。用蒙特卡洛模拟实验比较GRM-AHM的几种归类法的归准率（属性模式归准率和单个属性的平均判准率）。结果发现,新归类法的归准率与AHM中的方法A差不多,但比方法B高很多;随着被试作答失误率的提高,它们的归准率都有所下降。在归类精度和简单性方面,GRM-AHM都比Bolt等(2004)提出的多级评分融合模型（Fusion Model）好     

Do learned irrelevance and perseveration play a role during discrimination learning?

In both healthy participants and various patient populations, performance on attentional set-shifting tasks has been found to be affected by learned irrelevance and/or perseveration. The present study examined whether or not these processes also play a role during the initial discrimination learning phase of those tasks. To this end, participants first solved a multidimensional discrimination learning task. Thereafter, they underwent three types of shift of relevant and/or irrelevant stimulus attributes, which enabled the assessment of the separate contribution of perseveration and learned irrelevance to post-shift task performance. Subsequent correlational analyses revealed that the number of errors during initial discrimination learning was significantly correlated with the number of errors in the learned irrelevance-shift but not the perseveration-shift. This suggests that processes underlying learned irrelevance in post-shift task performance also play a significant role during initial discrimination learning. The implications of these results for interpreting deficits in shifting are discussed.     

基于项目节点的多级评分模型的统一

多级评分项目由于可以提供更多关于被试的信息而被广泛的使用。本文首先通过引用一个多级评分的数学试题, 给出了项目节点这一概念。假设被试在项目节点上的正确反应概率为二参数逻辑斯蒂模型之下, 本文通过分析三种不同类型的多级评分项目, 得出了三个评分模型, 其中一个和等级反应模型在形式上是一样的。鉴于我国目前考试测量所使用的多级评分项目的形式, 可以运用本文所述的项目节点的方法将项目评分模型统一提出。     

Stochastic Ordering Of the Latent Trait by the Sum Score Under Various Polytomous IRT Models

The sum score is often used to order respondents on the latent trait measured by the test. Therefore, it is desirable that under the chosen model the sum score stochastically orders the latent trait. It is known that unlike dichotomous item response theory (IRT) models, most polytomous IRT models do not imply stochastic ordering. It is unknown, however, (1) whether stochastic ordering is often or rarely violated and (2) whether violations yield a serious problem for practical data analysis. These are the central issues of this paper. First, some unanswered questions that pertain to polytomous IRT models implying stochastic ordering were investigated. Second, simulation studies were conducted to evaluate stochastic ordering in practical situations. It was found that for most polytomous IRT models that do not imply stochastic ordering, the sum score can be used safely to order respondents on the latent trait.The author would like to thank Klaas Sijtsma for commenting on earlier drafts of this paper.     

Invited Review—Survey of Value‐Focused Thinking: Applications,Research Developments and Areas for Future Research

Criteria are the central focus of multi‐criteria decision analysis. Many authors have suggested using our values (or preferences) to define the criteria we use to evaluate alternatives. Value‐focused thinking (VFT) is an important philosophy that advocates a more fundamental view of values in our decision making in our private and professional lives. VFT proponents advocate starting first with our values and then using our values to create decision opportunities, evaluate alternatives and finally develop improved alternatives. It has been 20 years since VFT was first introduced by Ralph Keeney. This paper surveys the VFT literature to provide a comprehensive summary of the significant applications, describe the main research developments and identify areas for future research. We review the scope and magnitude of VFT applications and the key developments in theory since VFT was introduced in 1992 and found 89 papers written in 29 journals from 1992 to 2010. We develop about 20 research questions that include the type of article (application, theory, case study, etc.), the size of the decision space (which, when given, ranged from $200K to billions of dollars), the contribution documented in the article (application benefits) and the research contributions (categorized by preferences, uncertainties and alternatives). After summarizing the answers to these questions, we conclude the paper with suggestions for improving VFT applications and potential future research. We found a large number of significant VFT applications and several useful research contributions. We also found an increasing number of VFT papers written by international authors. Copyright © 2012 John Wiley & Sons, Ltd.     

阈下特征框架效应心理机制初探

特征框架效应是指当分别以积极框架和消极框架来描述给定情景中的某一事物或事件的关键特征时,一般情况下人们更偏好以积极框架来描述其关键特征的事物或事件的现象.本研究通过2个实验验证了阈下特征框架效应的存在及其产生机制.实验一发现,阈下特征框架效应只在有时间压力下才会产生.实验二发现,被试对与框架类型一致的词汇判断的反应时比对与框架类型不一致的词汇判断的反应时更快,这在一定程度上说明特征框架效应的产生是因为框架激活了与框架一致性的信息,起到了语义启动的作用,从而导致被试判断任务的评定值偏向该框架类型.     

认知诊断框架下多级评分题目的DIF检测方法及其应用

本文对多级计分认知诊断测验的DIF概念进行了界定,并通过模拟实验以及实证研究对四种常见的多级计分DIF检验方法的适用性进行理论以及实践性的探索。研究结果表明：四种方法均能对多级计分认知诊断中的DIF进行有效的检验,且各方法的表现受模型的影响不大;相较于以总分为匹配变量,以KS为匹配变量时更利于DIF的检测;以KS为匹配变量的LDFA方法以及以KS为匹配变量的曼特尔检验方法在检测DIF题目时有着最高的检验力。