Psychometric properties of the Danish MCMI-I translation

A translation of the MCMI-I has been in use in Denmark for some years. An untested assumption in the interpretation of the pattern of test results is that the psychometric characteristics of the Danish and American versions are similar. The purpose of this study was to evaluate the psychometric properties of the questionnaire by using traditional psychometric analysis techniques on the results of a sample consisting of 423 patients and 179 normal controls. Coefficient alpha was calculated for the 20 clinical subscales of the test and the Danish results were strikingly similar to the original coefficients reported by Millon. Furthermore, factor analysis of the subscales showed a factor structure very similar to American findings, and it is concluded that the psychometric properties of the Danish MCMI are not significantly different from the original.     

Association coefficients of identity and proportionality for metric scales

Zegers' (1986) chance-corrected coefficients of association are derived by alternative methods. A different definition of chance correction is used. It is shown that our correction and that of Zegers are identical for large samples. Three possible assumptions for the derivation of metric coefficients are examined. The first, variable reflection, formulated by Zegers and ten Berge (1985), leads to coefficients that require chance-correction. Two other assumptions, zero covariance and covariance reflection, are proposed and it is shown that the latter two assumptions lead directly to coefficients of identity and proportionality that do not require chance correction (i.e., are identical to the Zegers, 1986, corrected coefficients).We are indebted to Robyn M. Dawes, Carnegie-Mellon University, for stimulating our interest in this project, and for helpful suggestions.     

测验同质性系数及其区间估计

在决定将多维测验分数合并成测验总分时, 应当考虑测验同质性。如果同质性太低, 合成总分没有什么意义。同质性高低可以用同质性系数来衡量。用来计算同质性系数的模型是近年来受到关注的双因子模型(既有全局因子又有局部因子), 测验的同质性系数定义为测验分数方差中全局因子分数方差所占的比例。本文用Delta法推导出计算同质性系数的标准误公式, 进而计算其置信区间。提供了简单的计算同质性系数及其置信区间的程序。用一个例子说明如何估计同质性系数及其置信区间, 通过模拟比较了用Delta法和用Bootstrap法计算的置信区间, 发现两者差异很小。     

High-pressure electrical resistivity behaviour of nanocrystalline perovskite (La,Sr)(Mn,Fe)O3

We report here the electrical resistivity of nanocrystalline perovskite-structured La–Sr manganites as a function of pressures up to 8?GPa, at room temperature. The nanocrystalline perovskite manganites were prepared by the sol–gel technique and found to have crystallite sizes of 12–18?nm. The pressure dependence of the electrical resistivity shows a first-order phase transition at 0.66(2)?GPa and a subtle phase transition between 3.5 and 3.8?GPa. The first-order transition at 0.66?GPa can be related to the transition from localized-electron to band magnetism.     

Effect of Reward Contingency on Intrinsic Motivation

The effects of type of task and type of reward contingency on intrinsic motivation for subsequent task performance were investigated. The sample consisted of 50 young Indian adults. The results showed a significantly greater amount of free choice time spent on a heuristic task rather than an algorithmic task. The verbal measures of intrinsic motivation yielded a significant effect of reward contingency with greater intrinsic motivation under performance-contingent reward, followed by task-contingent and no-reward conditions, respectively. The results support self-determination and competence processes as important mediators of intrinsic motivation.     

非正态分布测量数据对克隆巴赫信度α系数的影响

对“增大被试群体异质性,能提高测验信度”观点提出质疑。讨论非正态分布测量数据的偏度对克隆巴赫信度α系数的影响,以模拟方法验证Box—COX正态化变换对信度的高估现象,进而给出对克隆巴赫信度α系数估计的改进方法。     

多维测验分数的报告与解释:基于双因子模型的视角

在心理、行为、管理和教育等社科领域,经常使用多维测验。本文评介并比较了各种多维测验的测量模型;总结了基于双因子模型计算得到的统计指标;根据不同研究目的,提出了两个兼顾简洁性和精确性的多维测验分析流程。作为例子,在马基雅维利主义人格量表的研究中,通过双因子模型分析了如何报告、解释多维测验分数以及如何利用多维建模进行后续分析。     

Amount of General Factor Saturation in the Beck Anxiety Inventory Responses of Outpatients with Anxiety Disorders

To ascertain the amount of general factor saturation underlying the Beck Anxiety Inventory (BAI; Beck and Steer 1993) responses of anxious patients, the BAI was administered to 525 outpatients with DSM-IV-TR anxiety disorders. McDonald’s omega_hierarchical(ω _h) (Zinbarg et al. 2005) was calculated to estimate the amount of general factor saturation; ω _h is the amount of variance explained by a general (second-order) factor underlying a scale composed of multiple items divided by the total amount of variance explained by all of the scale’s items. An iterated principal-factor analysis was first performed and followed by a Schmid-Leiman transformation to determine the amount of variance explained by the second-order factor. ω _h was 0.70 [95% CI 0.66–0.72], and this value was discussed as supporting the current practice of summing the 21 symptom ratings of the BAI to estimate the overall severity of self-reported anxiety.

概化理论研究及应用前景

1972年,Cronbach和他的同事们提出概化理论之后,概化理论在行为与心理测量领域得到了广泛的应用,较之经典测量理论,它的优势逐渐地显露：(1)测量的多种误差来源可以在同一个分析中分别估计;(2)可以指导决策者选择最优测量方案;(3)提供可靠性系数：概化系数(G系数)和依存性指标(φ系数)用于不同的决策任务;(4)排除了严格平行测验的假设。概化理论以它的精确性和可藏性受到了信度测量领域研究者们的青睐,本文旨在对概化理论的基本框架、产生、发展及应用前景进行详细论述。     

关联规则方法在心理测量中的应用

探讨了当相关系数对变量之间相关程度不易进行度量或者无法度量的情况下,是否可以采用关联规则方法来考察变量之间的相互关系。从理论上分析了关联规则方法在心理测量中应用的可能性,并运用Apriori算法挖掘频繁项集,对来华留学生跨文化适应性数据进行了分析,将关联规则方法分析结果与Pearson相关、Spearman相关结果进行比较,发现当使用关联规则方法进行相互关系分析时,能够更加简易、有效地获得变量间的相关规则。研究结果表明关联规则方法可以获得变量各个水平具体的支持度和置信度,推导出多变量之间的相关规则,并且对于处理非线性的多变量以及多分类变量的相关是非常有用的方法。