When a Psychometric Advance Falls in the Forest

Borsboom (2006) attacks psychologists for failing to incorporate psychometric advances in their work, discusses factors that contribute to this regrettable situation, and offers suggestions for ameliorating it. This commentary applauds Borsboom for calling the field to task on this issue and notes additional problems in the field regarding measurement that he could add to his critique. It also chastises Borsboom for occasionally being unnecessarily perjorative in his critique, noting that negative rhetoric is unlikely to make converts of offenders. Finally, it exhorts psychometricians to make their work more accessible and points to Borsboom, Mellenbergh, and Van Heerden (2003) as an excellent example of how this can be done. I wish to thank Frank Schmidt for his help in preparing this paper. Requests for reprints should be sent to la-clark@uiowa.edu.     

目标追求的入世、出世心理测评初探

假设入世、出世心理是相对独立的成分,编制了相应的问卷。探索性因素分析和验证性因素分析表明,入世分量表由“拼搏精神”和“在乎结果”因素组成,出世分量表由“平常心”和“低要求”因素组成。该问卷具有较好的信度和效度。     

关于评价中心若干问题的探讨

本文对评价中心的定义、评价中心与心理测验的比较、评价中心的预测效度以及影响评价中心成功实施的原因等问题,进行了综合分析和初步探讨,从而帮助人们能够正确地认识评价中心,有效地使用评价中心;同时,也为人们对评价中心的研究和改革,提供了理论和实践方面的依据。     

Reconsidering reconsent: Threats to internal and external validity when participants reconsent after debriefing

We overwhelmingly utilize (partially) informed consent for, and debriefing of, human research participants. Also common is the practice of reconsent, particularly where changes in study protocols (or in participants themselves) occur midstream – participants consent again to remaining in the project or to having their data included. Worryingly under-discussed is post-debriefing reconsent, wherein participants can withdraw their data after learning more fully of the study's goals and methods. Yet, major ethics bodies in Canada, the United States and the United Kingdom promote such practice, with vague and potentially problematic guidelines. Here, the author provides examples involving such reconsent practice, highlighting potentially serious problems that are scientific (e.g. threats to internal and external validity) and ethical (i.e. to the participant, their peers, the researcher and society) in nature. Particularly, problematic is the introduction, by design, of unknowable bias in our research findings. For example, highly prejudiced participants could withdraw data from a discrimination study after learning of the study's hypotheses and goals. The practice may arguably contradict an Open Science goal of increasing research transparency. This call for discussion about the direction of psychological science methods aims to engage a broader discussion in the research community.     

应征公民心理选拔的人格评估

应征公民人格评估的目的是对精神分裂症病前人格特征进行预测性评价。采取定性与定量两种研究方法编制征兵专用人格问卷,并对其进行指标验证。结果发现：①量表应答得分区间可较好地区分正常新兵与精神分裂症被试;②二阶因子分析将人格分量表部分划分为3个维度8个因子;③精神分裂症患者人格分量表分数各指标均显著高于正常新兵;④总预测符合率和预测合格符合率均在98%以上,预测淘汰符合率最低为70.13%。上述结果表明,本研究编制的中国士兵人格问卷（CSPQ）具有良好的信度和效度,适用于我国应征公民心理检测以及我军士兵的人格测试     

不同学业成就中学生创造性思维的差异研究

本研究采用创造力量表对990名不同学业成就中学生的创造性思维进行研究,结果发现:(1)高学业成就组中学生的创造性思维、发散思维和聚合思维显著高于低学业成就组.(2)高低学业成就组中学生创造性思维、发散思维和聚合思维表现出年级差异性.(3)高低学业成就组中学生的创造性思维都随着年级发展而上升,但是两组的高峰期却出现差异,高学业成就组创造力高峰出现在初二和高二,而低学业成就组出现在高一.(4)发散思维的发展趋势上,高学业成就组呈现出曲折上升的趋势,而低学业成就组整体的发展趋势平稳.(5)高低学业成就组中学生聚合思维的发展趋势相似,都呈现出在初中阶段上升到高中阶段趋于稳定,并且在高中阶段两组的水平接近.     

法则关系方法在研究中的应用及其问题思考

重要概念之间的理论关系或合理关系的连锁系统被称作是一个法则关系。法则关系包括原有模型、前因变量或结果变量。应用结构方程模型来检验法则关系中各组成部分之间的法则有效性,从而对原模型作进一步修正,该方法能够更加科学地揭示出多变量间的因果关系。文中还提到了应用法则关系方法时需要注意的问题,并指出未来的研究应当着重于进一步构建其理论框架、借助其来有效地开发和验证测量工具、以及探索该方法效度的影响因素等方向。     

物质主义价值观量表在大学生群体中的修订

以449名大学本科生为被试,对物质主义价值观量表（MVS）进行中文版的修订,考察其信效度指标。探索性因素分析发现,修订后的量表具有与原量表基本相同的三个维度;验证性因素分析结果显示,MVS的三因素结构拟合较好;MVS的内部一致性信度为0．792,重测信度为0．830;MVS的效标关联效度良好。表明修订后的MVS具有较好的心理测量学属性,可作为测量我国大学生物质主义价值观的工具。     

群集化儿童问题行为核查表的中文版修订

将Frick等人的群集化儿童问题行为核查表（the clustering child behaviors checklist,CCBCL）修订为中文版自我报告问卷、教师评定问卷和同伴评定问卷。以529名初一学生为被试进行间卷修订。结果发现：Frick等人提出的外化问题行为框架适合研究中国儿童青少年的外化问题行为,探索性因子分析证明中文版问卷的因子结构与原文卷一致;自我报告、教师评定和同伴评定版问卷的内部一致性系数、分半系数和重测信度系数都在可接受的范围内,且自我报告版问卷中财物损害的各信度系数较低;自我报告结果与同伴拒绝的相关不显著。而教师评定和同伴评定的结果均与同伴拒绝相关显著;同伴评定问卷的各信度系数及通过其获得的测量结果与同伴拒绝的相关性均高于其它两种问卷。