More efficient parameter estimates for factor analysis of ordinal variables by ridge generalized least squares

Data in psychology are often collected using Likert‐type scales, and it has been shown that factor analysis of Likert‐type data is better performed on the polychoric correlation matrix than on the product‐moment covariance matrix, especially when the distributions of the observed variables are skewed. In theory, factor analysis of the polychoric correlation matrix is best conducted using generalized least squares with an asymptotically correct weight matrix (AGLS). However, simulation studies showed that both least squares (LS) and diagonally weighted least squares (DWLS) perform better than AGLS, and thus LS or DWLS is routinely used in practice. In either LS or DWLS, the associations among the polychoric correlation coefficients are completely ignored. To mend such a gap between statistical theory and empirical work, this paper proposes new methods, called ridge GLS, for factor analysis of ordinal data. Monte Carlo results show that, for a wide range of sample sizes, ridge GLS methods yield uniformly more accurate parameter estimates than existing methods (LS, DWLS, AGLS). A real‐data example indicates that estimates by ridge GLS are 9–20% more efficient than those by existing methods. Rescaled and adjusted test statistics as well as sandwich‐type standard errors following the ridge GLS methods also perform reasonably well.     

Inverse effectiveness, multisensory integration, and the bodily self: Some statistical considerations

A recent report in Consciousness and Cognition provided evidence from a study of the rubber hand illusion (RHI) that supports the multisensory principle of inverse effectiveness (PoIE). I describe two methods of assessing the principle of inverse effectiveness (‘a priori’ and ‘post-hoc’), and discuss how the post-hoc method is affected by the statistical artefact of ‘regression towards the mean’. I identify several cases where this artefact may have affected particular conclusions about the PoIE, and relate these to the historical origins of ‘regression towards the mean’. Although the conclusions of the recent report may not have been grossly affected, some of the inferential statistics were almost certainly biased by the methods used. I conclude that, unless such artefacts are fully dealt with in the future, and unless the statistical methods for assessing the PoIE evolve, strong evidence in support of the PoIE will remain lacking.     

运用均数与协方差结构模型侦查项目功能差异

阐释了运用多组均数与协方差结构(MACS)模型侦查多级反应项目的一致性与非一致性项目功能差异(DIF)的原理与程序, 以道德自我概念量表DIF的侦查进行示例, 并对该方法进行了评价。与项目反应理论比照, MACS采用系统的、迭代的方式利用修正指数来侦查DIF, 并提供多个拟合指数协同评价模型拟合;与标准验证性因素分析相较, MACS不仅能侦查非一致性DIF, 而且能侦查一致性DIF。运用MACS侦查DIF是一种值得推荐的方法。     

Correcting Too Much or Too Little? The Performance of Three Chi-Square Corrections

This simulation study investigates the performance of three test statistics, T₁, T₂, and T₃, used to evaluate structural equation model fit under non normal data conditions. T₁ is the well-known mean-adjusted statistic of Satorra and Bentler. T₂ is the mean-and-variance adjusted statistic of Sattertwaithe type where the degrees of freedom is manipulated. T₃ is a recently proposed version of T₂ that does not manipulate degrees of freedom. Discrepancies between these statistics and their nominal chi-square distribution in terms of errors of Type I and Type II are investigated. All statistics are shown to be sensitive to increasing kurtosis in the data, with Type I error rates often far off the nominal level. Under excess kurtosis true models are generally over-rejected by T₁ and under-rejected by T₂ and T₃, which have similar performance in all conditions. Under misspecification there is a loss of power with increasing kurtosis, especially for T₂ and T₃. The coefficient of variation of the nonzero eigenvalues of a certain matrix is shown to be a reliable indicator for the adequacy of these statistics.     

Divergence Behaviour of the Successive Geometric Mean Method of Pairwise Comparison Matrix Generation for a Multiple Stage,Multiple Objective Optimization Problem

This paper focuses on the divergence behaviour of the successive geometric mean (SGM) method used to generate pairwise comparison matrices while solving a multiple stage, multiple objective (MSMO) optimization problem. The SGM method can be used in the matrix generation phase of our three‐phase methodology to obtain pairwise comparison matrix at each stage of an MSMO optimization problem, which can be subsequently used to obtain the weight vector at the corresponding stage. The weight vectors across the stages can be used to convert an MSMO problem into a multiple stage, single objective (MSSO) problem, which can be solved using dynamic programming‐based approaches. To obtain a practical set of non‐dominated solutions (also referred to as Pareto optimal solutions) to the MSMO optimization problem, it is important to use a solution approach that has the potential to allow for a better exploration of the Pareto optimal solution space. To accomplish a more exhaustive exploration of the Pareto optimal solution space, the weight vectors that are used to scalarize the MSMO optimization problem into its corresponding MSSO optimization problem should vary across the stages. Distinct weight vectors across the stages are tied directly with distinct pairwise comparison matrices across the stages. A pairwise comparison matrix generation method is said to diverge if it can generate distinct pairwise comparison matrices across the stages of an MSMO optimization problem. In this paper, we demonstrate the SGM method's divergence behaviour when the three‐phase methodology is used in conjunction with an augmented high‐dimensional, continuous‐state stochastic dynamic programming method to solve a large‐scale MSMO optimization problem. Copyright © 2013 John Wiley & Sons, Ltd.     

Brain lateralization of complex movement: Neuropsychological evidence from unilateral stroke

Complex movement (CM) refers to the representation of a goal-oriented action and is classified as either transitive (use of tools) or intransitive (communication gestures). Both types of CM have three specific components: temporal, spatial, and content, which are subdivided into specific error types (SET). Since there is debate regarding the contribution of each brain hemisphere for the types of CM, our objective was to describe the brain lateralization of components and SET of transitive and intransitive CM. We studied 14 patients with a left hemisphere stroke (LH), 12 patients with a right hemisphere stroke (RH), and 16 control subjects. The Florida Apraxia Screening Test-Revised (FAST-R, Rothi et al., 1988) was used for the assessment of CM. Both clinical groups showed a worse performance than the control group on the total FAST-R and transitive movement scores (p < 0.001). Failures in Spatial and Temporal components were found in both clinical groups, but only LH patients showed significantly more Content errors (p < 0.01) than the control group. Also, only the LH group showed a higher number of errors for intransitive movements score (p = 0.017), due to lower scores in the content component, compared to the control group (p = 0.04). Transitive and intransitive CMs differ in their neurocognitive representation; transitive CM shows a bilateral distribution of its components when compared to intransitive CM, which shows a preferential left hemisphere representation. This could result from higher neurocognitive demands for movements that require use of tools, compared with more automatic communication gestures.     

Occupational differences in work engagement: A longitudinal study among eight occupational groups in Norway

The aim of the present study was to explore occupational differences in the experience of engagement both with regard to differences in the level of work engagement as well as in the predicting value of different antecedent variables. Multigroup latent mean analysis was performed on eight different occupational groups in Norway (lawyers, physicians, nurses, teachers, church ministers, bus drivers, and people working in advertising and information technology; N = 3,475). Tests for factorial invariance supported the use of the Oldenburg Burnout Inventory scale across occupational groups and that the latent means were comparable across the groups. Results indicated significant occupational differences in the experience of vigor and dedication. The lawyers reported the most vigor and the church ministers the most dedication. Least vigor was reported among the teachers and the advertising group reported to be least dedicated. Cross‐lagged multigroup structural equation modeling (SEM) analysis suggested there are different processes behind the development of engagement across occupations. Visualization of how some occupations cluster or differ from each other is important as it might engender theory building and further hypotheses testing.     

Testing the Physical Activity Self-Definition Model among low-active adults participating in a physical activity intervention

Seeing oneself as a physically active person is one of the strongest predictors of physical activity behaviour and self-regulatory strategies. Determining whether and how physical activity self-perceptions can be stimulated may help low-active individuals who do not see themselves as a physically active person become more active. Cross-sectional research has tested the Physical Activity Self-Definition (PASD) model among active samples; longitudinal studies among low-active adults have yet to be done. The purpose of this study was to test the predictive power of the PASD model among low-active adults over a 16-week physical activity intervention. Participants completed surveys of validated questionnaires in-person at baseline (pre-intervention) and at 16-weeks (end of intervention) at one of two primary care facilities. The final sample included 119 low-active adults. Partial least squares-structural equation modeling indicated that the original model had small-medium predictive power (Q² = 0.22; SRMR = 0.13 [0.05, 0.07]; RMSE = 1.13; MAE = 0.9; BIC = 1348.40). Two paths were added in the revised model (perceived wanting—PASD; perceived ability—perceived commitment), which explained an additional 4% and 5% of the variance in perceived commitment (R² = 0.62 [0.48, 0.72]) and PASD (R² = 0.74 [0.64, 0.80]; all p’s < 0.001), respectively. The revised model had medium predictive power (Q² = 0.25; SRMR = 0.11 [0.05, 0.06]; RMSE = 1.1; MAE = 0.87; BIC = 1332.84) All path coefficients remained positive and significant at p ≤ .001. Among low-active adults, perceived wanting and perceived ability may be more salient when engaging in physical activity and regarding themselves as a physically active person. Findings may support practitioners and health care professionals in designing physical activity interventions to foster PASD among low-active adult populations.     

Testing attitudes,social desirability and behavioral regulations as moderators of implicit-explicit exercise discrepancies: A replication study in Iranian students

IntroductionThere is mixed evidence regarding moderators of the relationship between implicitly and explicitly measured constructs. This study examined this issue with evaluations of exercise relative to health or appearance in a sample of Iranian adolescents. Investigaing automatic associations and reflective constructs might provide important implications for developing interventions targeting exercise behavior in a population with different cultural, social, and economic barriers to exercise than European or North American adolescents.MethodsParticipants were 471 students enrolled in grades 9 to 12, of whom 269 (54.9%) were female. Behavioral regulations, explicit attitudes, and social desirability were investigated as possible moderators. All students completed questionnaire measures of physical activity behavior, attitudes, social desieality and behavioral regulation. They also completed two Go/No Go Association tasks to measure automatic associations of exercise with health and apperarance.ResultsAttitude was a significant moderator of discrepancies between automatic associations of exercise with health and health motives. Interjected regulation moderated implicit-explicit appearance discrepancies. Participants with low social desirability and negative automatic appearance associations had the highest appearance motivation.ConclusionsParticipants focused on the potential of achieving positive health outcomes of exercise rather than physical appearance outcomes, which differs from research from other countries. It is possible that, health was strongly linked with exercise by the pervasive societal pressures in Iran to have a healthy body.     

Comparative Analysis of Work Values Across Four Nations

Work values are important factors in career counseling and development. The purpose of this study was to test the measurement invariance of Terada's ( 2009 ) Vocational Values Inventory, as well as compare the means of 5 latent variables across 4 nations (Japan, South Korea, the United States, and Germany) using structural equation modeling. Results indicated that the assumptions of configural, metric, and intercept invariance were satisfied across the 4 nations. A comparison of the means of the 5 latent variables indicated that job security, an extrinsic work value, was perceived as the most important work value across the 4 nations.