主客体互倚性的成对模式及其检验

主客体互倚性模型（APIM）以成对数据作为分析单元,是成对数据分析的重要方法。基于APIM中的主体效应与客体效应,Kenny等人论述了APIM的四种成对模式：主体模式、客体模式、对偶模式与对比模式。通过定义客体效应与主体效应的比值为k,Kenny等人提出了定量指标,以分析主客体互倚性的成对模式。成对模式分析有助于简化数据结果、检验理论模型、解析群体互动过程。虽然k值的检验程序较为复杂,但APIM成对模式及k值的提出,对于成对数据分析,特别是家庭研究与组织心理学研究具有重要意义。     

Time pressure and group performance: Exploring underlying processes in the Attentional Focus Model

The Attentional Focus Model (Karau & Kelly, 1992) predicts that time pressure should lead group members to focus on a restricted range of task-relevant cues and to adopt task completion as their major interaction objective. Although this prediction has been supported in several studies (e.g., Karau & Kelly, 1992; Kelly, Jackson, & Hutson-Comeaux, 1997; Kelly & Karau, 1999; Parks & Cowlin, 1995), the exact processes that underlie information restriction have not been specified. We propose that two processes are involved. Specifically, the restriction of information may occur because time pressure affects the way in which information is initially encoded or attended to before a group enters its decision-making phase, or because group members filter out what they judge to be less important information during group discussion and decision making. We assessed both of these processes within a decision-making experiment where time pressure was manipulated prior to learning information that would be used to perform a task. Recall of information learned prior to group discussion did not differ by time pressure condition, arguing against an encoding process. In contrast, interaction data demonstrated information restriction by group members, arguing for a filtering process. Thought listings collected 1 min into the group discussion in some groups also supported a filtering process, as did questionnaire data. Implications of these findings for the Attentional Focus Model and for effective group decision making are discussed.     

Likert量表分析中不同IRT模型的有效性

5级Likert量表可直接分析,也可以转化为3级评分,或转化为2级评分。前二者可以采用等级IRT模型,后者可以采用2级IRT模型。研究表明2级IRT模型中的2参数模型是最适合的模型。多级评分模型与数据拟合也很好,而且等级越多测量精度越大。     

A Multivariate Associative Finite Growth Mixture Modeling Approach Examining Adolescent Alcohol and Marijuana Use

Theoretical and empirical substance use development research suggests that adolescent populations are not homogenous and can often be separated into subpopulations characterized by qualitatively different patterns of substance use development. This paper demonstrates the application of a multivariate associative finite latent growth mixture modelling approach to examine heterogeneity in patterns of adolescent alcohol and marijuana use and the influence of age, gender, parent, and peer substance use. Substance use problem outcomes were also examined. Participants were male and female adolescents (N = 1,044) ranging in age from 11 to 17 years at the first assessment (Mean age = 14.47; SD = 1.95). Individuals were 45% female and 82% Caucasian. Using growth mixture methodology, a 7-class model captured distinct simultaneous alcohol and marijuana use patterns over a 3-year period. Findings highlight the importance of examining subgroups of adolescent substance use, rather than focusing only on single samples.     

I feel,therefore you act: Intrapersonal and interpersonal effects of emotion on negotiation as a function of social power

We examine how emotion (anger and happiness) affects value claiming and creation in a dyadic negotiation between parties with unequal power. Using a new statistical technique that analyzes individual data while controlling for dyad-level dependence, we demonstrate that anger is helpful for powerful negotiators. They feel more focused and assertive, and claim more value; the effects are intrapersonal, insofar as the powerful negotiator responds to his or her own emotional state and not to the emotional state of the counterpart. On the other hand, effects of emotion are generally not intrapersonal for low-power negotiators: these negotiators do not respond to their own emotions but can be affected by those of a powerful counterpart. They lose focus and yield value. Somewhat surprisingly, the presence of anger in the dyad appears to foster greater value creation, particularly when the powerful party is angry. Implications for the negotiation and power literatures are discussed.     

A developmental typology of adolescent personality

The purpose of the current study is to examine whether Block's personality types (i.e. Resilients, Undercontrollers and Overcontrollers) are replicable as developmental trajectories. We applied a Latent Class Growth Analysis (LCGA) framework to five‐ annual‐wave data on a sample of early to middle adolescents (n = 923). Our results showed that Block's Resilients, Undercontrollers and Overcontrollers are indeed replicable as developmental trajectories across adolescence. These developmental types were related to problem behaviour in a similar way as types found in studies using cross‐sectional data. As such, Resilients reflected low levels of problem behaviour, Undercontrollers had high levels of delinquency and Overcontrollers had high levels of depression. Implications and suggestions for further research are discussed. Copyright © 2009 John Wiley & Sons, Ltd.     

文本阅读中时间信息加工特点及二阶段模型的建构

时间信息的加工一直是当代阅读心理研究的热点。以往研究对时间信息表征的性质、维度特征、提取机制等问题存在诸多争论, 主要表现为强印象假设与场景理论的冲突。我们认为, 文本阅读中时间信息的加工包括两个阶段：加工时间词以及低水平的更新情景模型阶段和通达与整合先前信息阶段, 强印象假设和场景理论争论的实质在于前者解释的是加工的第一个阶段, 而后者解释的是加工的第二个阶段, 因此, 总是难以调和。本项目首先在更高的理论层面上将二者加以整合, 然后采用经典的认知心理学研究方法与现代的眼动研究技术相结合的方法, 通过对文本阅读中时间信息表征性质、加工的维度特征、提取过程的激活与抑制特征等问题的探讨, 为本项目提出的时间信息加工的二阶段模型提供行为实验与眼动实验证据, 从而解决该领域中强印象假设与场景理论的长期争论。本项目的理论构想与研究技术对文本阅读其他问题的探讨具有重要的方法论意义。     

移情在道德强度对企业道德决策影响中的作用

摘 要 本研究将一种重要的道德情绪--移情引入问题权变模型中,以探讨移情因素在道德强度对企业道德决策影响中的作用。本研究基于情景研究法,随机选取256名MBA学生为被试,要求被试基于所提供情景中假想参与者的行为,回答随后测量道德决策(道德识别、道德判断和道德意图)、移情反应、主观道德强度及移情特质的问题。结果发现,1）移情反应在道德强度对企业道德决策的影响中具有中介作用;2）移情关怀特质会通过影响主观道德强度和移情反应而作用于企业道德决策;3)结果大小、社会舆论和效应可能性对道德判断和道德意图的影响机制各有特点。     

Measuring model flexibility with parameter space partitioning: an introduction and application example

A primary criterion on which models of cognition are evaluated is their ability to fit empirical data. To understand the reason why a model yields a good or poor fit, it is necessary to determine the data-fitting potential (i.e., flexibility) of the model. In the first part of this article, methods for comparing models and studying their flexibility are reviewed, with a focus on parameter space partitioning (PSP), a general-purpose method for analyzing and comparing all classes of cognitive models. PSP is then demonstrated in the second part of the article in which two connectionist models of speech perception (TRACE and ARTphone) are compared to learn how design differences affect model flexibility.