scholarly journals Synthesizing Indirect Effects in Mediation Models with Meta-Analytic Methods

2020 ◽  
Author(s):  
Mike W.-L. Cheung
Keyword(s):  
Indirect Effects ◽  
Structural Equation ◽  
Meta Analysis ◽  
Underlying Mechanism ◽  
Equation Modeling ◽  
Outcome Variable ◽  
Correlation Matrices ◽  
Independent Variable ◽  
Research Questions ◽  
Mediation Models

A mediator is a variable that explains the underlying mechanism between an independent variable and a dependent variable. The indirect effect indicates the effect from the predictor to the outcome variable via the mediator. In contrast, the direct effect represents the effect of the predictor on the outcome variable after controlling for the mediator. A single study rarely provides enough evidence to answer research questions in a particular domain. Replications are generally recommended as the gold standard to conduct scientific research. When a sufficient number of studies have been conducted addressing similar research questions, a meta-analysis can be used to synthesize the findings of those studies. The main objective of this paper is to introduce two approaches to integrating studies using mediation analysis. The first approach involves calculating standardized indirect effects and direct effects and conducting a multivariate meta-analysis on those effect sizes. The second approach uses meta-analytic structural equation modeling to synthesize correlation matrices and fit mediation models on the average correlation matrix. We illustrate these procedures on a real dataset using the R statistical platform. This paper closes with some further directions for future studies.ind

scholarly journals Meta-Analytic Structural Equation Modeling

2021 ◽  
Author(s):  
Mike W.-L. Cheung
Keyword(s):  
Structural Equation Modeling ◽  
Correlation Matrix ◽  
Structural Equation ◽  
Structural Equation Models ◽  
Meta Analysis ◽  
Equation Modeling ◽  
Average Correlation ◽  
Correlation Matrices ◽  
Factor Analytic Model ◽  
Special Cases

Meta-analysis and structural equation modeling (SEM) are two popular statistical models in the social, behavioral, and management sciences. Meta-analysis summarizes research findings to provide an estimate of the average effect and its heterogeneity. When there is moderate to high heterogeneity, moderators such as study characteristics may be used to explain the heterogeneity in the data. On the other hand, SEM includes several special cases, including the general linear model, path model, and confirmatory factor analytic model. SEM allows researchers to test hypothetical models with empirical data. Meta-analytic structural equation modeling (MASEM) is a statistical approach combining the advantages of both meta-analysis and SEM for fitting structural equation models on a pool of correlation matrices. There are usually two stages in the analyses. In the first stage of analysis, a pool of correlation matrices is combined to form an average correlation matrix. In the second stage of analysis, proposed structural equation models are tested against the average correlation matrix. MASEM enables researchers to synthesize researching findings using SEM as the research tool in primary studies. There are several popular approaches to conduct MASEM, including the univariate-r, generalized least squares, two-stage SEM (TSSEM), and one-stage MASEM (OSMASEM). MASEM helps to answer the following key research questions: (a) Are the correlation matrices homogeneous? (b) Do the proposed models fit the data? (c) Are there moderators that can be used to explain the heterogeneity of the correlation matrices? The MASEM framework has also been expanded to analyze large datasets or big data with or without the raw data.

Meta-Analytical SEM: Equivalence Between Maximum Likelihood and Generalized Least Squares

2018 ◽  
Vol 43 (6) ◽  
pp. 693-720
Author(s):  
Ke-Hai Yuan ◽  
Yutaka Kano
Keyword(s):  
Maximum Likelihood ◽  
Least Squares ◽  
Correlation Matrix ◽  
Structural Equation ◽  
Structural Model ◽  
Meta Analysis ◽  
Generalized Least Squares ◽  
Equation Modeling ◽  
Correlation Matrices ◽  
Latent Constructs

Meta-analysis plays a key role in combining studies to obtain more reliable results. In social, behavioral, and health sciences, measurement units are typically not well defined. More meaningful results can be obtained by standardizing the variables and via the analysis of the correlation matrix. Structural equation modeling (SEM) with the combined correlations, called meta-analytical SEM (MASEM), is a powerful tool for examining the relationship among latent constructs as well as those between the latent constructs and the manifest variables. Three classes of methods have been proposed for MASEM: (1) generalized least squares (GLS) in combining correlations and in estimating the structural model, (2) normal-distribution-based maximum likelihood (ML) in combining the correlations and then GLS in estimating the structural model (ML-GLS), and (3) ML in combining correlations and in estimating the structural model (ML). The current article shows that these three methods are equivalent. In particular, (a) the GLS method for combining correlation matrices in meta-analysis is asymptotically equivalent to ML, (b) the three methods (GLS, ML-GLS, ML) for MASEM with correlation matrices are asymptotically equivalent, (c) they also perform equally well empirically, and (d) the GLS method for SEM with the sample correlation matrix in a single study is asymptotically equivalent to ML, which has being discussed extensively in the SEM literature regarding whether the analysis of a correlation matrix yields consistent standard errors and asymptotically valid test statistics. The results and analysis suggest that a sample-size weighted GLS method is preferred for combining correlations and for MASEM.

scholarly journals Explaining Effects of Transformational Leadership on Teachers’ Cooperative Professional Development through Structural Equation Model and Phantom Model Approach

2021 ◽  
Vol 13 (19) ◽  
pp. 10888
Author(s):  
Woonsun Kang
Keyword(s):  
Professional Development ◽  
Transformational Leadership ◽  
Indirect Effects ◽  
Self Efficacy ◽  
Structural Equation ◽  
Positive Impact ◽  
Transformative Leadership ◽  
Underlying Mechanism ◽  
Equation Modeling ◽  
Multiple Mediation

This study aims to explore strategies for promoting Korean teachers’ cooperative professional development in the context of Education for Sustainable Development (ESD). To this end, this study aimed to provide empirical evidence that may help in explaining the mechanism through which transformative leadership affects cooperative professional development. Based on research purposes, the author constructed a serial multiple mediation model that incorporates self-efficacy and attitudes as an underlying mechanism that explains transformational leaders’ positive impact on cooperative professional development related to ESD. A total of 203 valid cases were used to test the hypothesized model. Furthermore, the author constructed structural equation modeling (SEM) and phantom models specifying the specific indirect effects of transformative leadership on cooperative professional development. In addition, bias-corrected bootstrapped point estimates for the specific indirect effects were calculated. The data were analyzed using SPSS Ver. 25.0 and AMOS Ver. 26.0. The findings strongly supported predictive links among transformational leadership, self-efficacy in ESD teachers’ attitudes toward ESD, and cooperative professional development. Therefore, findings provided greater insight into transformational leadership and ESD research and revealed mechanisms through which transformational leadership works in the ESD contexts. Finally, the major findings were discussed to encourage teachers to participate in cooperative professional development.

scholarly journals Meta-Analytic Structural Equation Modeling

2020 ◽  
Author(s):  
Mike W.-L. Cheung
Keyword(s):  
Structural Equation Modeling ◽  
Correlation Matrix ◽  
Structural Equation ◽  
Structural Equation Models ◽  
Meta Analysis ◽  
Equation Modeling ◽  
Average Correlation ◽  
Correlation Matrices ◽  
Factor Analytic Model ◽  
Special Cases

Meta-analysis and structural equation modeling (SEM) are two popular statistical models in the social, behavioral, and management sciences. Meta-analysis summarizes research findings to provide an estimate of the average effect and its heterogeneity. When there is non-trial heterogeneity, moderators such as study characteristics may be used to explain the heterogeneity in the data. On the other hand, SEM includes several special cases, including the general linear model, path model, and confirmatory factor analytic model. SEM allows researchers to test hypothetical models with empirical data. Meta-analytic structural equation modeling (MASEM) is a statistical approach combining the advantages of both meta-analysis and SEM for fitting structural equation models on a pool of correlation matrices. There are usually two stages in the analyses. In the first stage of analysis, a pool of correlation matrices is combined to form an average correlation matrix. In the second stage of analysis, proposed structural equation models are tested against the average correlation matrix. MASEM enables researchers to synthesize researching findings using SEM as the research tool in primary studies. There are several popular approaches to conduct MASEM, including the univariate-r, generalized least squares, two-stage SEM (TSSEM), and one-stage MASEM (OSMASEM). MASEM helps to answer the following key research questions: (1) Are the correlation matrices homogeneous? (2) Do the proposed models fit the data? (3) Are there moderators that can be used to explain the heterogeneity of the correlation matrices? The MASEM framework has also been expanded to analyze large datasets or big data with or without the raw data.

scholarly journals Some Reflections on Combining Meta-Analysis and Structural Equation Modeling

2018 ◽  
Author(s):  
Mike W.-L. Cheung
Keyword(s):  
Social Sciences ◽  
Structural Equation Modeling ◽  
Structural Equation ◽  
Meta Analysis ◽  
Equation Modeling ◽  
Synthesis Methods ◽  
Software Packages ◽  
Research Questions ◽  
Personal Reflections ◽  
New Research

Meta-analysis and structural equation modeling (SEM) are two of the most prominent statistical techniques employed in the behavioral, medical, and social sciences. They each have their own well-established research communities, terminologies, statistical models, software packages, and journals (Research Synthesis Methods and Structural Equation Modeling: A Multidisciplinary Journal). In this paper, I will provide some personal reflections on combining meta-analysis and SEM in the forms of meta-analytic SEM (MASEM) and SEM-based meta-analysis. The critical contributions of Becker (1992), Shadish (1992), and Viswesvaran and Ones (1995) in the early development of MASEM are highlighted. Another goal of the paper is to illustrate how meta-analysis can be extended and integrated with other techniques to address new research questions such as the analysis of Big Data. I hope that this paper may stimulate more research development in the area of combining meta-analysis and SEM.

Current Methods for Meta-Analysis

2007 ◽  
Vol 215 (2) ◽  
pp. 90-103 ◽  
Author(s):  
Ralf Schulze
Keyword(s):  
Fixed Effects ◽  
Structural Equation ◽  
Meta Analysis ◽  
Effect Sizes ◽  
Equation Modeling ◽  
Random Effects Model ◽  
New Methods ◽  
New Findings ◽  
Multivariate Procedures ◽  
Research Questions

Abstract. The bulk of conceptual and statistical developments as well as applications of meta-analysis have been published in the last 30 years. The methods for meta-analysis continue to be refined and new methods are applied to new types of research questions and data. Such current approaches, issues, and developments prevalent in the behavioral sciences are presented, reviewed, and discussed in this paper. The areas that are covered include: the fixed effects and random effects model of meta-analysis, new findings concerning effect sizes and their statistical properties, the comparison of different meta-analytic approaches, and multivariate procedures for meta-analysis. The latter also covers the stepwise combination of meta-analysis and structural equation modeling (MASEM).

Direct and Indirect Effects of Interdental Hygiene in a Clinical Trial

2008 ◽  
Vol 87 (11) ◽  
pp. 1037-1042 ◽  
Author(s):  
Y.-K. Tu ◽  
M. Jackson ◽  
M. Kellett ◽  
V. Clerehugh
Keyword(s):  
Indirect Effects ◽  
Structural Equation ◽  
Analysis Of Covariance ◽  
Equation Modeling ◽  
Outcome Variable ◽  
Direct And Indirect Effects ◽  
Direct Effects ◽  
Pocket Depth ◽  
Randomized Controlled ◽  
Probing Pocket Depth

Many randomized controlled trials (RCTs) in dental research test the efficacy of interventions on more than one outcome variable. Univariate methods, such as the t test or analysis of covariance, cannot evaluate the efficacy of interventions on multiple outcomes simultaneously. The aim of this study was to use structural equation modeling (SEM) to re-analyze a RCT, comparing the effects of pre-curved interdental brushes and flossing on probing pocket depth (PPD), plaque indices, and bleeding on probing (BOP) measured at baseline, intermediate, and final examinations. Results of SEM showed that the observed greater reduction in PPD and BOP in persons using interdental brushing than in those flossing is due mainly to the greater efficiency in plaque removal with the interdental brushes (indirect effect) rather than to the compression of the interdental papillae (direct effect). In contrast, smokers showed less BOP at baseline but also less improvement in BOP through direct effects.

scholarly journals Emotional Responses and Perceived Relative Harm Mediate the Effect of Exposure to Misinformation about E-Cigarettes on Twitter and Intention to Purchase E-Cigarettes among Adult Smokers

2021 ◽  
Vol 18 (23) ◽  
pp. 12347
Author(s):  
Jessica Liu ◽  
Caroline Wright ◽  
Olga Elizarova ◽  
Jennifer Dahne ◽  
Jiang Bian ◽  
...  
Keyword(s):  
Indirect Effects ◽  
Structural Equation ◽  
Emotional Responses ◽  
Affective Responses ◽  
Equation Modeling ◽  
Web Based ◽  
Intention To Purchase ◽  
The Relationship ◽  
Different Levels ◽  
Mediation Models

There is a gap in knowledge on the affective mechanisms underlying effects of exposure to health misinformation. This study aimed to understand whether discrete emotional responses and perceived relative harm of e-cigarettes versus smoking mediate the effect of exposure to tweets about the harms of e-cigarettes on Twitter and intention to purchase e-cigarettes among adult smokers. We conducted a web-based experiment in November 2019 among 2400 adult smokers who were randomly assigned to view one of four conditions of tweets containing different levels of misinformation. We fitted mediation models using structural equation modeling and bootstrap procedures to assess the indirect effects of exposure to tweets through perceived relative harm of e-cigarettes and six discrete emotions. Our findings support that exposure to tweets about harms of e-cigarettes influence intention to purchase e-cigarettes through perceived relative harm, discrete emotional responses, and serially through emotional responses and perceived relative harm. Feeling worried, hopeful, and happy mediated the effects of condition on intention to purchase e-cigarettes. Feeling scared, worried, angry, and hopeful mediated the effects serially through perceived relative harm. Affective responses and perceived relative harm following exposure to misinformation about e-cigarette harm may mediate the relationship with intention to purchase e-cigarettes among adult smokers.

Risk Factors and Mediators of Suicidal Ideation Among Korean Adolescents

Crisis ◽  
2018 ◽  
Vol 39 (1) ◽  
pp. 4-12 ◽  
Author(s):  
Yi Jin Kim ◽  
Sung Seek Moon ◽  
Jang Hyun Lee ◽  
Joon Kyung Kim
Keyword(s):  
High School Students ◽  
Suicidal Ideation ◽  
Structural Equation ◽  
Equation Modeling ◽  
Outcome Variable ◽  
Suicidal Behaviors ◽  
Thwarted Belongingness ◽  
Perceived Burdensomeness ◽  
Korean Adolescents ◽  
Related Stress

Abstract. Background: A significant number of Korean adolescents have suicidal ideations and it is more prevalent among adolescents than any other age group in Korea. Aims: This study was conducted to attain a better understanding of the contributing factors to suicidal ideation among Korean adolescents. Method: We recruited 569 high school students in Grades 10 and 11 in Pyeongtaek, Korea. The Beck Scale for Suicidal Ideation was used to measure suicidal ideation as the outcome variable. The Interpersonal Needs Questionnaire, the Beck Hopelessness Scale, the School Related Stress Scale, the Olweus Bully/Victim Questionnaire, and the Youth Risk Behavior Surveillance questions were used to measure thwarted belongingness and perceived burdensomeness, hopelessness, school-related stress, bullying, and previous suicidal behaviors, respectively. Data analyses included descriptive statistics and structural equation modeling. Results: The findings suggest that perceived burdensomeness, hopelessness, school-related stress, and previous suicidal behaviors have significant direct effects on suicidal ideation. Hopelessness fully mediated the relation between thwarted belongingness and suicidal ideation, and partially mediated between perceived burdensomeness, school-related stress, and suicidal ideation. Conclusion: These findings provide more specific directions for a multidimensional suicide prevention program in order to be successful in reducing suicide rates among Korean adolescents.

Impulsivity and Aggression: A Meta-analysis Using the UPPS Model of Impulsivity

2019 ◽  
Author(s):  
Konrad Bresin
Keyword(s):  
Structural Equation Modeling ◽  
Sensation Seeking ◽  
Structural Equation ◽  
Meta Analysis ◽  
The Other ◽  
Equation Modeling ◽  
Negative Urgency ◽  
Two Stage ◽  
Trait Impulsivity

Trait impulsivity has long been proposed to play a role in aggression, but the results across studies have been mixed. One possible explanation for the mixed results is that impulsivity is a multifaceted construct and some, but not all, facets are related to aggression. The goal of the current meta-analysis was to determine the relation between the different facets of impulsivity (i.e., negative urgency, positive urgency, lack of premeditation, lack of perseverance, and sensation seeking) and aggression. The results from 93 papers with 105 unique samples (N = 36, 215) showed significant and small-to-medium correlations between each facet of impulsivity and aggression across several different forms of aggression, with more impulsivity being associated with more aggression. Moreover, negative urgency (r = .24, 95% [.18, .29]), positive urgency (r = .34, 95% [.19, .44]), and lack of premeditation (r = .23, 95% [.20, .26]) had significantly stronger associations with aggression than the other scales (rs < .18). Two-stage meta-analytic structural equation modeling showed that these effects were not due to overlap among facets of impulsivity. These results help advance the field of aggression research by clarifying the role of impulsivity and may be of interest to researchers and practitioners in several disciplines.

Sign in / Sign up

Export Citation Format

Share Document