结构方程模型统计检验力分析：原理与方法

结构方程模型是心理学、管理学、社会学等学科中重要的统计工具之一。然而, 大量使用结构方程模型的研究忽视了对该方法的统计检验力进行必要的分析和报告, 在一定程度上降低了这些研究的结果的证明效力。结构方程模型的统计检验力分析方法主要有Satorra-Saris法、MacCallum法与Monte Carlo法三类。其中Satorra-Saris法适用于备择模型清晰、检验对象相对简单、检验方法基于χ²分布的情形; MacCallum法适用于基于χ²分布的模型拟合检验且备择模型不明的情形; Monte Carlo法适用于检验对象相对复杂、采用模拟或重抽样方法进行检验的情形。在实际应用中, 研究者应当首先判断检验的目的、方法以及是否有明确的备择模型, 并根据这些信息选择具体的分析方法。     

Participatory and multiobjective development of watercourse regulation—creation of regulation alternatives from stakeholders' preferences

Multicriteria decision analysis (MCDA) methods have been recently applied in many environmental problems. Major aims have been to structure and to analyse multifaceted and complex problems, to compare incommensurable impacts and to clarify the preference order of alternatives. Most MCDA tools concentrate on aiding the process of choice among alternatives. The choice usually occurs at the end of the decision‐making process, but MCDA tools can also assist earlier in the process. In this article, we present a new MCDA‐based method in order to create watercourse regulation alternatives, which meet the objectives of stakeholders. The method is comprised of three elements: (1) framework for the planning and learning process partly based on the Image Theory, (2) analysis and evaluation of ecological, social and economic impacts of regulation, and (3) visual interactive Excel implementation of value‐tree analysis (REGAIM‐model). We show how the method was applied in a complex watercourse regulation development project in Finland. Altogether 36 face‐to‐face computer‐aided interviews were undertaken with the REGAIM model with representatives of different stakeholder groups. We present the main results of the interviews and discuss how these interviews supported generation of new watercourse regulation alternatives. We also describe the advantages of the new approach in the participatory watercourse management, and discuss the applicability of Image Theory in the watercourse regulation context. Copyright © 2006 John Wiley & Sons, Ltd.     

基于语义差异量表的心理疾病外显与内隐污名的结构及关系

采用语义差异法编制心理疾病污名语义差异量表,用于测量外显污名,并用语义差异量表的词对作为单类内隐联想测验的材料,测量内隐污名,用这两类材料研究外显污名与内隐污名的结构及关系。结果表明：外显污名的三维度模型具有良好的拟合度（χ²/df=2.99, RMSEA=0.075, CFI=0.918, TLI=0.902）;内隐污名的三维度模型也具有良好的拟合度（χ²/df=2.272, RMSEA=0.071, CFI=0.857, TLI=0.833）,且与外显污名的结构相同,都由认知评价、情感体验、行为反应三维度构成;心理疾病污名的外显认知评价与内隐认知评价存在显著相关（r=0.436）,外显污名与内隐污名的其它维度之间均不存在显著相关。     

On the importance of distinguishing between within- and between-subject effects in intransitive intertemporal choice

Most probabilistic paired comparison models treat inconsistent choices as caused by independent and random errors in the pairwise judgments. In this paper, we argue that this assumption is too restrictive for the analysis of paired comparison data obtained from multiple judges when transitivity violations are systematic. We present a new framework that contains the random error assumption as a special case but also allows for systematic changes in an option's utility assessments over the pairwise comparisons. Accounting for both between- and within-judge sources of variability, we demonstrate in an application on intertemporal choice that the proposed framework can capture systematic transitivity violations as well as individual taste differences.     

The confluence model: birth order as a within-family or between-family dynamic?

The confluence model explains birth-order differences in intellectual performance by quantifying the changing dynamics within the family. Wichman, Rodgers, and MacCallum (2006) claimed that these differences are a between-family phenomenon--and hence are not directly related to birth order itself. The study design and analyses presented by Wichman et al. nevertheless suffer from crucial shortcomings, including their use of unfocused tests, which cause statistically significant trends to be overlooked. In addition, Wichman et al. treated birth-order effects as a linear phenomenon thereby ignoring the confluence model's prediction that these two samples may manifest opposing results based on age. This article cites between- and within-family data that demonstrate systematic birth-order effects as predicted by the confluence model. The corpus of evidence invoked here offers strong support for the assumption of the confluence model that birth-order differences in intellectual performance are primarily a within-family phenomenon.     

An Investigation of the Sample Performance of Two Nonnormality Corrections for RMSEA

The root mean square error of approximation (RMSEA) is a popular fit index in structural equation modeling (SEM). Typically, RMSEA is computed using the normal theory maximum likelihood (ML) fit function. Under nonnormality, the uncorrected sample estimate of the ML RMSEA tends to be inflated. Two robust corrections to the sample ML RMSEA have been proposed, but the theoretical and empirical differences between the 2 have not been explored. In this article, we investigate the behavior of these 2 corrections. We show that the virtually unknown correction due to Li and Bentler (2006) Li, L. and Bentler, P. M. 2006. “Robust statistical tests for evaluating the hypothesis of close fit of misspecified mean and covariance structural models.”. In UCLA Statistics Preprint #506. Los Angeles: University of California..  [Google Scholar], which we label the sample-corrected robust RMSEA, is a consistent estimate of the population ML RMSEA yet drastically reduces bias due to nonnormality in small samples. On the other hand, the popular correction implemented in several SEM programs, which we label the population-corrected robust RMSEA, has poor properties because it estimates a quantity that decreases with increasing nonnormality. We recommend the use of the sample-corrected RMSEA with nonnormal data and its wide implementation.     

Aparametric investigation of multiletter matches

Proctor (1981) presented a theoretical framework that distinguishes factors contributing to disparities in time to classify physical-same, name-same, and different letter pairs as a function of three variables: method of presentation (simultaneous vs. successive), case relationship (samecase vs. opposite-case pairs), and blocking (blocked vs. mixed presentation of same-case and opposite-case pairs). He also argued that these variables were critical in multIletter matches, and performed a between-study comparison of existing data to support his contention. Because comparison across studies is always a tenuous process and because the absence of several relevant conditions precluded a complete analysis of predictions, a within-experiment manipulation of the three relevant variables was desirable. The present study reports such an experiment. In general, the factorial manipulation of variables supported predictions of Proctor’s framework and indicated that many phenomena of multiletter matching, including the widely studied fast-“same” phenomenon, are attributable primarily to differences in the rate at which component letter pairs are matched.     

贝叶斯多组比较——渐近测量不变性

在行为科学研究领域中,检验测量工具的测量不变性是进行群体差异比较的前提。目前,多组验证性因子分析(多组CFA)方法被广泛用于检验测量不变性,但是它对跨组等值的限制过于严格,在实际应用中常常存在大量局限。贝叶斯渐近测量不变性方法基于贝叶斯思想的优良特性,放宽了传统多组CFA方法对跨组差异的严格限制,避免了传统方法的问题,具有较高的应用价值。文章详细介绍了贝叶斯渐近测量不变性方法的原理及优势,同时通过实例展示了渐近测量不变性方法在Mplus软件中的具体分析过程。     

The Problem with Having Two Watches: Assessment of Fit When RMSEA and CFI Disagree

The root mean square error of approximation (RMSEA) and the comparative fit index (CFI) are two widely applied indices to assess fit of structural equation models. Because these two indices are viewed positively by researchers, one might presume that their values would yield comparable qualitative assessments of model fit for any data set. When RMSEA and CFI offer different evaluations of model fit, we argue that researchers are likely to be confused and potentially make incorrect research conclusions. We derive the necessary as well as the sufficient conditions for inconsistent interpretations of these indices. We also study inconsistency in results for RMSEA and CFI at the sample level. Rather than indicating that the model is misspecified in a particular manner or that there are any flaws in the data, the two indices can disagree because (a) they evaluate, by design, the magnitude of the model's fit function value from different perspectives; (b) the cutoff values for these indices are arbitrary; and (c) the meaning of “good” fit and its relationship with fit indices are not well understood. In the context of inconsistent judgments of fit using RMSEA and CFI, we discuss the implications of using cutoff values to evaluate model fit in practice and to design SEM studies.     

The dynamics of choice in a changing world: Effects of full and partial feedback

We explored the dynamics of choice behavior while the values of the options changed, unannounced, several times. In particular, choice dynamics were compared when the outcome values of all available options were known (full feedback) and when the outcome value of only the chosen option was known (partial feedback). The frequency of change, the values of the options, and the difference between them were also manipulated. In an experiment with N = 427, we found that the patterns of choices were different for the two levels of feedback. Whereas behavior in the full-feedback condition showed a tendency to switch choices following a missed opportunity—replicating previous findings—the behavior in the partial-feedback condition was different. It was sensitive to the outcome value of the chosen option in comparison to some memory of the last-experienced outcome value of the unchosen option. However, the comparison of these two values influenced choice behavior only when the outcome of the currently chosen option was satisfactory and the last outcome of the unchosen one was not. As expected, the other manipulated variables (change frequency, the options’ values, and the difference between them) had no effect on the dynamics of behavior.     

Structural alignment facilitates the noticing of differences

High-similarity concept pairs that elicit many commonalities also elicit many related differences (Gentner & Markman, 1994; A. B. Markman & Gentner, 1993a, 1993b, 1996; A. B. Markman & Wisniewski, 1997). This finding has been used to support the claim that the comparison process is one of structural alignment. However, it is possible that the difference advantage results from some other property of high-similarity pairs, such as a greater number of stored differences. The present experiments demonstrate that the comparison process itself leads to the greater psychological availability of differences. In three experiments, participants listed commonalities for word pairs and then listed differences under a time pressure for these old pairs and new pairs. In Experiment 1, participants listed more differences for old than for new pairs, consistent with the claim that the comparison process facilitates noticing differences. In Experiment 2, we showed that the difference-listing advantage is specific to the comparison process: Mere coprocessing of the pairs (specifically, providing thematic relations) does not facilitate, and in fact appears to inhibit, difference listing. In Experiment 3, pairs with deeper common systems elicited a larger number of specific alignable differences than did pairs with shallow sets of commonalities. Overall, the results support the structural alignment claim that the comparison process promotes the noticing of both commonalities and related differences.     

The Multiplicative AHP,SMART and ELECTRE in a Common Context

This paper presents a comparative study of three popular methods for multicriteria decision analysis based on a particular model of human preferential judgement. Since decisions are invariably made within a given context, we model relative preferences as ratios of increments or decrements in an interval on an axis of desirability. Next we sort the ratio magnitudes into a small number of categories, represented by numerical values on a geometric scale. We explain why the analytic hierarchy process (AHP) and the French collection of ELECTRE methods, typically based on pairwise comparison methods, are concerned with categories of ratio magnitudes, whereas the simple multiattribute rating technique (SMART) essentially uses orders of magnitude of these ratios. This phenomenon provides a common basis for the analysis of the methods in question and for a cross-validation of their results. We illustrate the approach via a well-known case study, the choice of a location for a nuclear power plant. We conclude by discussing the scope of the comparative study. © 1997 John Wiley & Sons, Ltd.     

The Influence of Number of Categories and Threshold Values on Fit Indices in Structural Equation Modeling with Ordered Categorical Data

This study examines the unscaled and scaled root mean square error of approximation (RMSEA), comparative fit index (CFI), and Tucker–Lewis index (TLI) of diagonally weighted least squares (DWLS) and unweighted least squares (ULS) estimators in structural equation modeling with ordered categorical data. We show that the number of categories and threshold values for categorization can unappealingly impact the DWLS unscaled and scaled fit indices, as well as the ULS scaled fit indices in the population, given that analysis models are misspecified and that the threshold structure is saturated. Consequently, a severely misspecified model may be considered acceptable, depending on how the underlying continuous variables are categorized. The corresponding CFI and TLI are less dependent on the categorization than RMSEA but are less sensitive to model misspecification in general. In contrast, the number of categories and threshold values do not impact the ULS unscaled fit indices in the population.     

The influence of social comparison on career decision-making: Vocational identity as a moderator and regret as a mediator

This study investigated the influence of social comparison on career choice certainty and its potential mechanisms: regret as a mediator and vocational identity as a moderator. Before the formal experiment, 30 pairs of vocational values representing typical conflicts in career decision-making for Chinese university students were obtained. The formal experiment adopted a single-factor (social comparison VS no comparison) between-subject design with vocational identity as an independent covariate. Ninety-eight junior and senior undergraduate students and graduate students in a university in China were invited to participate in the computer-controlled experiment, which involved vocational identity assessment, social comparison manipulation, and analogue career-choice scenario tests. Path analysis showed that: (a) Social comparison significantly and negatively predicted career choice certainty; (b) Regret partially mediated the effect of social comparison on career choice certainty; and (c) Vocational identity did not moderate the path between social comparison and regret, but significantly moderated the negative effect of regret on career choice certainty. These results indicated that in the collectivistic Chinese culture, individuals' career development trajectories may not be totally independent and are subject to influences by other people's choices, while emotion of regret and vocational identity development all play significant roles in this intricate process.     

A game-theoretic comparison of the utilitarian and maximin rules of social choice

I will characterize the utilitarian and maximin rules of social choice game-theoretically. That is, I will introduce games whose solutions are the utilitarian and maximin distributions respectively. Then I will compare the rules by exploring similarities and differences between these games. This method of comparison has been carried out by others. But I characterize the two rules using games that involve bargaining within power structures. This new characterization better highlights the ethical differences between the rules.     

Nonmetric tests of ratio vs. subtractive theories of stimulus comparison

Theories of stimulus comparison were tested by examining ordinal properties of data obtained with six scaling tasks. Subjects judged simple “ratios” or “differences” of stimulus pairs constructed from a factorial design. In four additional tasks, the same judges also compared relations between pairs of stimulus pairs, judging “ratios of ratios,” “ratios of differences,” “differences of ratios,” and “differences of differences.” The data were consistent with a subtractive theory, which asserts that two stimuli are compared by subtraction, regardless of the task, but that judges can compare two stimulus differences by either a ratio or a difference. All six tasks could be related by the subtractive theory using a single set of scale values. Other simple theories, including the theory that “ratio” judgments can be represented by a ratio model, could not reproduce the six rank orders (of the six sets of data) using a single set of scale values.     

Inferring the structure of latent class

Present optimization techniques in latent class analysis apply the expectation maximization algorithm or the Newton-Raphson algorithm for optimizing the parameter values of a prespecified model. These techniques can be used to find maximum likelihood estimates of the parameters, given the specified structure of the model, which is defined by the number of classes and, possibly, fixation and equality constraints. The model structure is usually chosen on theoretical grounds. A large variety of structurally different latent class models can be compared using goodness-of-fit indices of the chi-square family, Akaike’s information criterion, the Bayesian information criterion, and various other statistics. However, finding the optimal structure for a given goodness-of-fit index often requires a lengthy search in which all kinds of model structures are tested. Moreover, solutions may depend on the choice of initial values for the parameters. This article presents a new method by which one can simultaneously infer the model structure from the data and optimize the parameter values. The method consists of a genetic algorithm in which any goodness-of-fit index can be used as a fitness criterion. In a number of test cases in which data sets from the literature were used, it is shown that this method provides models that fit equally well as or better than the models suggested in the original articles.     

Hierarchical modeling of paired comparison data

The method of paired comparisons belongs to a small group of techniques that provide explicit information about the consistency of individual and aggregated choices. This article investigates the link between the individual- and group-level judgments by extending R. D. Luce's (1959) model, which was originally developed for individual choice behavior, to a mixed-effects paired comparison model. It is shown that standard multilevel software for binary data can be used to estimate the model. The interpretation of the paired comparison parameters and statistical model tests are discussed in detail. An extensive analysis of an experimental study illustrates the usefulness of a hierarchical approach in modeling multiple pairwise judgments.     

The self‐choice effects on memory and individual differences in emotional intelligence

The present study examined the self‐choice effects on intentional memory and the relation between individual differences in emotional intelligence, and the size of the self‐choice effects. Word pairs were supplied to the participants in either a self‐choice or forced‐choice condition. In the self‐choice condition the participants were asked to choose one of the words that elicited more pleasant or more unpleasant episodes, and to remember the chosen word. In the forced‐choice condition one of the paired words was underlined to signify that it was the word to remember. For participants who chose a word that elicited more pleasant episodes, self‐choice effects were observed in pleasant‐pleasant and pleasant‐unpleasant word pairs. However, for participants who were asked to choose a word that elicited more unpleasant episodes, the effects were found in unpleasant‐unpleasant and pleasant‐unpleasant word pairs. The size of the self‐choice effects was determined by the level of emotional intelligence.