Multilevel Modeling with Correlated Effects

When there exist omitted effects, measurement error, and/or simultaneity in multilevel models, explanatory variables may be correlated with random components, and standard estimation methods do not provide consistent estimates of model parameters. This paper introduces estimators that are consistent under such conditions. By employing generalized method of moments (GMM) estimation techniques in multilevel modeling, the authors present a series of estimators along a robust to efficient continuum. This continuum depends on the assumptions that the analyst makes regarding the extent of the correlated effects. It is shown that the GMM approach provides an overarching framework that encompasses well-known estimators such as fixed and random effects estimators and also provides more options. These GMM estimators can be expressed as instrumental variable (IV) estimators which enhances their interpretability. Moreover, by exploiting the hierarchical structure of the data, the current technique does not require additional variables unlike traditional IV methods. Further, statistical tests are developed to compare the different estimators. A simulation study examines the finite sample properties of the estimators and tests and confirms the theoretical order of the estimators with respect to their robustness and efficiency. It further shows that not only are regression coefficients biased, but variance components may be severely underestimated in the presence of correlated effects. Empirical standard errors are employed as they are less sensitive to correlated effects when compared to model-based standard errors. An example using student achievement data shows that GMM estimators can be effectively used in a search for the most efficient among unbiased estimators. This research was supported by the National Academy of Education/Spencer Foundation and the National Science Foundation, grant number SES-0436274. We thank the editor, associate editor, and referees for detailed feedback that helped improve the paper.     

The Role of Theory in Testing Hypothesized Interactions: An Example from the Research on Escalation of Commitment1

Theory should guide the choice of analytic strategy in testing research hypotheses. This thesis is developed primarily with respect to the testing of interaction hypotheses, with some discussion of the implications for testing hypotheses regarding main effects. Previously published data on the escalation bias (Schoorman, 1988) are re-analyzed using planned contrasts, two-way analysis of variance (ANOVA), and regression analyses using dummy and effect coded variables, with both simultaneous and hierarchical entry. Each analysis yields different results, which demonstrates that different analytic strategies can lead to conflicting interpretations regarding the theory of interest. The reasons for these differences and the implications of these findings for the choice of analytic strategies are discussed.     

The Relationship among Metacognitions,Attentional Control,and State Anxiety

Abstract

The present study explored the relationship among metacognitions, attentional control, and state anxiety. A convenience sample of 142 undergraduate students completed the Meta-Cognitions Questionnaire-30, the Attentional Control Scale, and the State subscale of the State–Trait Anxiety Inventory 3 weeks before end-of-year examinations. A cross-sectional design was adopted, and data analysis consisted of correlation and hierarchical regression analyses. Correlation analyses showed that three dimensions of metacognition (negative beliefs about thoughts concerning uncontrollability and danger, cognitive confidence, and beliefs about the need to control thoughts) were positively correlated with state anxiety. These same metacognitions were also found to be negatively correlated with attention shifting and, with the exception of cognitive confidence, attention focusing. Both attention focusing and attention shifting were found to be negatively correlated with state anxiety. A hierarchical regression analysis showed that negative beliefs about thoughts concerning uncontrollability and danger and attention focusing independently predicted state anxiety. Overall, these findings support the hypotheses and are consistent with the metacognitive theory of psychological dysfunction in that they show that metacognitions (in the form of negative beliefs about thoughts concerning uncontrollability and danger) and executive control (in the form of attention-focusing) individually contribute to state anxiety. The implications of these findings are discussed.     

Therapist Multicultural Competence, Asian American Participants' Cultural Values, and Counseling Process

Asian Americans drop out of mental health treatment at a high rate. This problem could be addressed by enhancing therapists' multicultural competence and by examining clients' cultural attitudes that may affect the counseling process. In the present study, we used a video analogue design with a sample of 113 Asian American college students to examine these possibilities. The result from a t test showed that the session containing therapist multicultural competencies received higher ratings than the session without therapist multicultural competence. In addition, correlational analyses showed that participant values acculturation was positively associated with participant ratings of counseling process, while the value of emotional self-control was negatively correlated. The results of a hierarchical multiple regression analysis did not support any interaction effects among the independent variables on counseling process. All of these findings could contribute to the field of multicultural competence research and have implications for therapist practices and training.     

Bayesian modeling of measurement error in predictor variables using item response theory

It is shown that measurement error in predictor variables can be modeled using item response theory (IRT). The predictor variables, that may be defined at any level of an hierarchical regression model, are treated as latent variables. The normal ogive model is used to describe the relation between the latent variables and dichotomous observed variables, which may be responses to tests or questionnaires. It will be shown that the multilevel model with measurement error in the observed predictor variables can be estimated in a Bayesian framework using Gibbs sampling. In this article, handling measurement error via the normal ogive model is compared with alternative approaches using the classical true score model. Examples using real data are given.This paper is part of the dissertation by Fox (2001) that won the 2002 Psychometric Society Dissertation Award.     

基因-环境交互作用理论模型及其检验方法

素质?压力模型、不同易感性模型和优势敏感性模型是当前有关基因?环境交互作用的三种代表性理论模型。素质?压力模型认为“近墨者黑”或“出淤泥而不染”, 优势反应敏感性模型认为“近朱者赤”, 而不同易感性模型则兼收并蓄, 认为某些个体近墨则黑, 近朱则赤。检验上述模型的现实有效性是当前基因?环境交互作用研究领域的热点问题之一。概观而言, 分组回归和分层回归是常用的传统检验方法, 显著性区域分析法和新参数回归模型法则是新近兴起的。未来研究需要进一步探索三种模型的领域特殊性、种族差异等问题, 检验这些模型的方法也有待改善。     

青少年主观幸福感与网络成瘾:自尊及自我控制的中介作用

以1551名初高中生为被试,采用问卷调查法探讨主观幸福感与网络成瘾的关系,以及自尊及自我控制在其中的中介效应。结果表明:(1)主观幸福感与自尊、自我控制能力之间显著正相关,且均与网络成瘾显著负相关;(2)青少年的主观幸福感水平显著负向影响其网络成瘾,而且还通过自尊和自我控制间接影响青少年的网络成瘾,并且间接效应更加明显;(3)在主观幸福感作用于网络成瘾的关系中,自尊及自我控制单独作用时的中介效应一般,但是当两者联合作用时中介效应大幅增加。     

Detection of interactions between a dichotomous moderator and a continuous predictor in moderated multiple regression with heterogeneous error variance

Moderated multiple regression (MMR) has been widely used to investigate the interaction or moderating effects of a categorical moderator across a variety of subdisciplines in the behavioral and social sciences. In view of the frequent violation of the homogeneity of error variance assumption in MMR applications, the weighted least squares (WLS) approach has been proposed as one of the alternatives to the ordinary least squares method for the detection of the interaction effect between a dichotomous moderator and a continuous predictor. Although the existing result is informative in assuring the statistical accuracy and computational ease of the WLS-based method, no explicit algebraic formulation and underlying distributional details are available. This article aims to delineate the fundamental properties of the WLS test in connection with the well-known Welch procedure for regression slope homogeneity under error variance heterogeneity. With elaborately systematic derivation and analytic assessment, it is shown that the notion of WLS is implicitly embedded in the Welch approach. More importantly, extensive simulation study is conducted to demonstrate the conditions in which the Welch test will substantially outperform the WLS method; they may yield different conclusions. Welch’s solution to the Behrens-Fisher problem is so entrenched that the use of its direct extension within the linear regression framework can arguably be recommended. In order to facilitate the application of Welch’s procedure, the SAS and R computing algorithms are presented. The study contributes to the understanding of methodological variants for detecting the effect of a dichotomous moderator in the context of moderated multiple regression. Supplemental materials for this article may be downloaded from brm.psychonomic-journals.org/content/supplemental.     

Moderating effect of self-determination in the relationship between Big Five personality and academic performance

Previous studies have shown that both personality and motivation are important factors in student academic performance. This study examined how the interactions between the Big Five personality traits and self-determination motivation orientations affect students' academic performance. The hypotheses were empirically tested using cross-sectional data collected from 249 primary school students in China. The correlation analysis found that self-determined motivation and four of the five personality traits (not emotional instability) were significantly positively related to academic performance in English. The hierarchical regression analysis revealed that, after controlling for gender, openness to new experience and conscientiousness both positively predicted English performance. Significant interaction effects were found between agreeableness and self-determined motivation, and between conscientiousness and self-determined motivation. However, conscientiousness and agreeableness only positively predicted academic performance when the student's self-determined motivation was low.     

Emotion work and job stressors and their effects on burnout

Abstract

This article reports research on emotion work, organizational as well as social variables as predictors of job burnout. In burnout research, high emotional demands resulting from interactions with clients are seen as a core characteristic of service jobs. However, these emotional demands were seldom measured in a direct manner. It was only recently that emotional demands were included in studies on burnout referring to the concept of emotion work (emotional labor). Emotion work is defined as the requirement to display organizationally desired emotions. A multi-dimensional concept of emotion work was used to analyze the relations of emotion work variables with organizational and social variables and their joint effect on burnout in five samples including employees working in children's homes, kindergartens, hotels, banks and call centers. Emotion work variables correlated with organizational stressors and resources. However, hierarchical multiple regression showed a unique contribution of emotion work variables in the prediction of burnout. Moreover, the analysis of interaction effects of emotional dissonance and organizational and social stressors showed that for service professionals, the coincidence of these stressors led to exaggerated levels of emotional exhaustion and depersonalization.     

Testing the Culture-Fit Hypothesis with religious beliefs: a multilevel analysis

This study tested the hypothesis that internalising cultural values that fit internalised Christian religious orientations would produce higher levels of well-being than internalising cultural values that are incongruous with one's religious orientation. Participants (n?=?409) completed self-report measures of cultural values (individualism and collectivism), religious orientation (quest and Christian orthodoxy) and well-being. A series of hierarchical regression analyses and hierarchical linear modelling revealed several Cultural Value X Religious Orientation interaction effects. Specifically, people whose cultural values did not cohere with their religious orientation experienced lower levels of well-being than those whose cultural values fit well with their religious orientation.     

基于两水平回归模型的调节效应分析及其效应量

使用多元回归法进行调节效应分析在社科领域已常有应用。简述了目前多元回归法的调节效应分析存在的不足,包括人为变换检验模型、自变量和调节变量区分不足、误差方差齐性的假设难以满足、调节效应量指标ΔR²没有直接测量调节变量对自变量与因变量关系的调节程度。比较好的方法是用两水平回归模型进行调节效应分析并使用相应的效应量指标。在介绍新方法和新效应量后,总结出一套调节效应的分析流程,通过一个例子来演示如何用Mplus软件进行两水平回归模型的调节效应及其效应量分析。最后讨论了两水平回归模型的调节效应分析的发展,包括稳健的调节效应分析、潜变量的调节效应分析、有调节的中介效应分析和有中介的调节效应分析等。     

Determination of the Regression Coefficients and Their Associated Standard Errors in Hierarchical Regression Analysis

Hierarchical regression analysis is potentially a very useful statistical technique for establishing the significance of sets of predictor variables. However, when a hierarchical analysis which is based on theory is performed, some estimation procedures for the regression coefficients and their associated standard errors are potentially inappropriate. Specifically, the hierarchical regression equations, the incremental or hierarchical tests, and the parameter estimation of this procedure may not correspond. This problem is investigated by the development of four approaches (simultaneous, stagewise, orthogonal, and hierarchical) of estimation to the analysis. For each method, regression, coefficient estimators and their standard errors are determined. By comparison of these approaches, the use of orthogonalized sets of predictor variates or a modification to a series of simultaneous analyses are recommended as the most sensible technique for a theory driven hierarchical analysis.     

Resilience and quality of life among empty nesters in China: the mediating role of the source of support

Abstract

This study aims to explore the correlations among resilience, social support, and quality of life (QOL) in empty nesters to comprehend whether various sources of social support indirectly affected the correlation between resilience and QOL. We used hierarchical multiple regression analysis and structural equation modeling method in this study. Resilience and support from both family members and friends significantly correlated with the QOL. Besides, both types of support played vital roles in the mediating effects on the correlation between resilience and QOL among empty nesters residing in China. Hence, the hypothesis proposed was confirmed partially. Furthermore, this study offers a comprehensive understanding of overall mental and physical health among empty nesters.     

新世纪20年国内心理统计方法研究回顾

新世纪头20年, 国内心理学11本专业期刊一共发表了213篇统计方法研究论文。研究范围主要包括以下10类(按论文篇数排序)：结构方程模型、测验信度、中介效应、效应量与检验力、纵向研究、调节效应、探索性因子分析、潜在类别模型、共同方法偏差和多层线性模型。对各类做了简单的回顾与梳理。结果发现, 国内心理统计方法研究的广度和深度都不断增加, 研究热点在相互融合中共同发展; 但综述类论文比例较大, 原创性研究论文比例有待提高, 研究力量也有待加强。     

道德推脱对员工道德决策的影响：德行领导的调节作用

运用问卷调查的方法,以819名员工为研究被试,探讨了德行领导在道德推脱影响道德决策过程中的调节作用。采用层次回归分析的方法研究发现：德行领导对道德推脱与道德识别、道德判断之间的关系具有调节作用;但德行领导不会对道德推脱与道德意图之间的关系产生调节作用,而会对道德意图产生显著的直接影响。     

Interactive effects of personality and separation as acculturation style on adolescent antisocial behaviour

The primary aim of this study was to assess the interaction between personality and acculturation variables in predicting antisocial behaviour among adolescent immigrants living in Spain. Previous studies have shown that the acculturation style referred to as separation (rejection of the host country's culture, and a strong desire to preserve the culture of the country of origin) was strongly related to antisocial behaviour among immigrants. A further aim of this study was to examine if the relationship between separation and antisocial behaviour was moderated by personality variables, in particular impulsivity, sensation-seeking, and personal competence. Self-reported data of a sample of 750 adolescent immigrants living in Galicia's and Madrid's regions (Spain) were gathered using previously validated scales. The results revealed that both separation and personality variables were significantly related to antisocial behaviour. Moreover, hierarchical regression analysis with the interaction terms entered found significant moderating effects i.e., the relationship between separation and antisocial behaviour was significantly amplified when impulsivity or sensation-seeking were high. The results underscore the need to examine the effects of personality, acculturation, and the interaction between both on antisocial behaviour among adolescent immigrants.     

The positive role of hope on the relationship between loneliness and unhappy conditions in Hungarian young adults: How pathways thinking matters!

ABSTRACT

In this study, we examined loneliness and hope components as predictors of unhappy conditions (viz., anxious symptoms, depressive symptoms, & suicidal ideation) in young adults. The sample was comprised of 489 Hungarian college students. Results of conducting hierarchical regression analyses indicated that loneliness and hope pathways (but not hope agency) were important unique predictors of anxious symptoms, depressive symptoms, and suicidal ideation. Moreover, in part, consistent with the notion that hope might buffer the negative effects of loneliness on unhappy conditions, evidence for a significant Loneliness × Hope Pathways interaction effect in predicting each of the three indices of unhappy conditions was found. In contrast, the Loneliness × Hope Agency interaction effect was not found to be significant. Some implications of the present findings for the study and treatment of unhappy conditions in adults are discussed.     

Investigation of the Contribution of Spirituality and Religiousness to Hedonic and Eudaimonic Well-Being in Iranian Young Adults

Studies that have used both spirituality and religiousness measures to predict mental well-being are virtually nonexistent. Furthermore, in almost all of the past studies, the distinction between hedonic and eudaimonic aspects of well-being has not been taken into account. To address these limitations, the present study sought to examine the relation between spirituality and religiousness and a rather comprehensive set of well-being scales in an Iranian Muslim sample. Participants were 292 undergraduates at the University of Tehran. Results of bivariate correlation analysis showed that all aspects of hedonic and eudaimonic well-being were positively correlated with all aspects of spirituality and religiousness. Results of the multiple regression analysis demonstrated that spirituality was a stronger predictor of well-being than religiousness. In order to further examine the unique contribution of spirituality and religiousness over each other and gender, hierarchical regression analysis was applied. Findings indicated that while spirituality was a significant predictor of well-being after partialling out the effects of gender and religiousness, religiousness did not add significantly to the prediction of well-being over and above the contribution of gender and spirituality. Implications of the results are discussed, as are study limitations and directions for further research.     

Emotion work and job stressors and their effects on burnout

Abstract This article reports research on emotion work, organizational as well as social variables as predictors of job burnout. In burnout research, high emotional demands resulting from interactions with clients are seen as a core characteristic of service jobs. However, these emotional demands were seldom measured in a direct manner. It was only recently that emotional demands were included in studies on burnout referring to the concept of emotion work (emotional labor). Emotion work is defined as the requirement to display organizationally desired emotions. A multi-dimensional concept of emotion work was used to analyze the relations of emotion work variables with organizational and social variables and their joint effect on burnout in five samples including employees working in children's homes, kindergartens, hotels, banks and call centers. Emotion work variables correlated with organizational stressors and resources. However, hierarchical multiple regression showed a unique contribution of emotion work variables in the prediction of burnout. Moreover, the analysis of interaction effects of emotional dissonance and organizational and social stressors showed that for service professionals, the coincidence of these stressors led to exaggerated levels of emotional exhaustion and depersonalization.