信度的再认识与信度概括化研究

本文首先对信度概念进行了明确,指出信度是评价测验结果可靠与否的一个指标,而不是测验工具的不变属性。针对测验结果的信度估计的可变性,介绍了上世纪末Vacha-Haase提出的信度概括化研究方法．即一种用来探索得分信度估计的可变性、并对引起变异的预测源进行探讨的一种元分析方法。最后通过对信度概括化研究手段的分析,指出信度概念的再认识与信度概括化研究将会给心理测验工作者带来新的启示。     

The effects on the predictive variance of a new subject’s score for a new test

Abstract: It is often required to predict the scores or their variations under interest. Ishii and Watanabe (2001) investigated, in the context of psychological measurement, the Bayesian predictive distribution of a new subject’s scores for tests and subjects’ scores for a new test. In this paper, the Bayesian posterior predictive distribution of a new subject’s scores for a new parallel test were considered. And the effects of the number of subjects, the number of the tests, and the test reliability were investigated. Then, it was found that, under assumptions that (co)variance parameters are known, the predictive variance of a new subject’s score for a new test was equal to the predictive variances of the new subject’s scores for the existent tests. It was also found that the effect of the number of subjects was relatively large and the effect of the number of tests was relatively small, when a new subject’s scores for existent tests were not observed.     

LGD在选拔企业中高级管理人才方面的实证研究

本研究利用在为某企业人才招聘活动中提供大型人才测评服务时所收集到的数据,分析研究了无领导小组讨论测评技术的信效度。结果表明,在LGD测评技术中,评价者经过系统培训后,其评分者信度较高,但在不同要素的评价方面存在一定的差异性;在效度方面,通过多种途径的数据检验都表明LGD是一种比较有效的人才测评技术。同时,本研究还得出结论．与个性品质方面的评价相比,LGD在评价个人的能力素质特征方面更有优势。另外,结合本次人才测评的实际操怍经验．研究者还提出了有关LGD测评实施过程的应用建议。     

自编《中小学生非智力因素教师评价量表》的信、效度研究

通过自行编制的《中小学生非智力因素教师评价量表》,以天津市4所中、小学校的4个年级365名学生为被试考察了该量表的质量。研究结果表明：(1)《中小学生非智力因素教师评价量表》具有较高的信度;(2)《中小学生非智力因素教师评价量表》所构想的非智力因素结构模型是完全可以接受的,即有较好的结构效度。     

Age, Gender, and Ethnicity Effects on Parent–Child Discrepancy Using Identical Item Measures

Objective assessment of child and adolescent behavioral and emotional symptoms is traditionally obtained from multiple sources. However, a substantial body of research indicates that parental and child reports provide significant amounts of contradicting diagnostic information. Although a large and growing body of research attempts to identify potential influences of discrepant reports, the current research improves upon previous research in three primary ways: using identical item measures, using expanded statistical analyses, and evaluating cultural influences on observed discrepancies. A total of 2,153 parent–child dyads completed ratings of child behavior and emotional functioning. Specifically, parents and children completed the Ohio Scales, an empirically supported, identical item measure. Generally, reporter agreement was greater than typically reported. Similar to previous research with clinical populations, parents reported greater levels of child problems than their children. While age was not associated with observed discrepancies, parents and daughters demonstrated greater discrepancies on fewer specific items while parents and sons demonstrated more pervasive yet less severe discrepancies. Additionally, Hispanic dyads demonstrated less discrepancy than did African American and Caucasian dyads independent of discrepancy analysis. Discrepancies must be measured using multiple statistical methods in order to understand patterns. Furthermore, discrepancy research must address key demographic factors (e.g., ethnicity, gender).     

Reliability Beyond Theory and Into Practice

The critical reactions of Bentler (2009, doi:), Green and Yang (2009a, doi:; 2009b, doi:), and Revelle and Zinbarg (2009, doi:) to Sijtsma’s (2009, doi:) paper on Cronbach’s alpha are addressed. The dissemination of psychometric knowledge among substantive researchers is discussed.     

多元概化理论在结构化面试中的应用研究

使用多元概化理论对由7名评委和34名考生组成的结构化面试数据进行了分析.结果表明:1.面试设置的5个维度中,仪表举止维度与其它4个维度之间的协方差相对较小;2.各个维度的概化系数都在0.69以上,其中,概化系数最小的是仪表举止维度,最大的是口语表达维度;3.通过增加评委人数可以提高面试信度,但增幅递减.     

Reliability and validity of mothers' developmental estimates for children between 4 and 41 months

The feasibility of mothers estimating their own children's development is explored in this paper. Construction, reliability and validity of a mothers' developmental inventory for children from 4 months to 41 months is described. The inventory consists of six subtests: Gross Motor, Fine Motor, Language Expression, Language Comprehension, Personal-Social Competence and Self Help. A preliminary version of the inventory was item analyzed on a sample of 105 mothers of 4 to 37 months old children. This reduced the inventory to 164 items with satisfactory reliability and age discrimination on half-year intervals. The list was cross-validated on a sample of 275 mothers, for which medians of alpha coefficients for the six subtests ranged from 0.69 to 0.96, with most (26 out of 36) coefficients being 0.77 or higher. Again, the age discrimination was satisfactory, although floor effects appeared on four subtests for children younger than 12 months and ceiling effects on two subtests for children older than 24 months.     

Beyond the qualitative paradigm: A framework for introducing diversity within qualitative psychology

The case for qualitative research in psychology is considered. We argue against the idea that qualitative research is merely a matter of technique or method, and question the utility of viewing it as a unitary paradigm. Rather, the links between epistemology, methodology, and method are explored within three theorized strands of qualitative inquiry, making reference to illustrative projects. Each strand is organized around a different approach to the issues of justifying and warranting psychological knowledge: (1) reliability and validity; (2) generativity and grounding; and (3) discourse and reflexivity. These are exemplified in Miles and Huberman's ‘data display’ model, Glaser and Strauss' method of ‘grounded theory’, and in various forms of ‘discourse’ analysis. Reflections upon points of contact between the three strands address two main issues: (1) rendering research publicly accountable; and (2) challenging relativism.     

CDCC中国儿童发展量表（3－6岁）的编制

中国儿童发展量表（３－６岁）是由张厚粲教授主持,北京师大和新疆师大等单位合作,参考国内外有关研究成果,结合我国特点编制而成。其内容包括语言、认知、社会认知和动作四个方面的１１个项目。量表长度适中,内容形式多样,易于施测;具有较高的信度和效度,正式通过了专家鉴定。几年的试用实践证明本量表是一个可靠而有效的测验工具。