scholarly journals Issues in Solving the Problem of Effect Size Heterogeneity in Meta-Analytic Structural Equation Modeling: A Commentary and Simulation Study on Yu, Downes, Carter, and O’Boyle (2016)

2017 ◽  
Author(s):  
Mike W.-L. Cheung
Keyword(s):  
Structural Equation Modeling ◽  
Structural Equation ◽  
Goodness Of Fit ◽  
Structural Equation Models ◽  
Parametric Bootstrap ◽  
Theoretical Models ◽  
Equation Modeling ◽  
Organizational Studies ◽  
Full Information ◽  
Correlation Matrices

This paper is in press at Journal of Applied Psychology.Abstract:Meta-analytic structural equation modeling (MASEM) is becoming increasingly popular for testing theoretical models from a pool of correlation matrices in management and organizational studies. One limitation of the conventional MASEM approaches is that the proposed structural equation models are only tested on the average correlation matrix. It remains unclear how far the proposed models can be generalized to other populations when the correlation matrices are heterogeneous. Recently, Yu, Downes, Carter, and O’Boyle (2016) proposed a full information MASEM approach to address this limitation by fitting structural equation models from the correlation matrices generated from a parametric bootstrap. However, their approach suffers from several conceptual issues and technical errors. In this study, we reran some of the simulations in Yu et al. by correcting all of the errors in their original studies. The findings showed that bootstrap credible intervals (CVs) work reasonably well, while test statistics and goodness-of-fit indices do not. We advise researchers on what they can and cannot achieve by applying the full information MASEM approach. We recommend fitting MASEM with the TSSEM approach, which works well for the simulation studies. If researchers want to inspect the heterogeneity of the parameters, they may use the bootstrap CVs from the full information MASEM approach. All of these analyses were implemented in the open-source R statistical platform; researchers can easily apply and verify the findings. This paper concludes with several future directions to address the issue of heterogeneity in MASEM.

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.

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.

On the Structure of Affect

2021 ◽  
Vol 229 (1) ◽  
pp. 24-37 ◽  
Author(s):  
Nadine Wedderhoff ◽  
Timo Gnambs ◽  
Oliver Wedderhoff ◽  
Tanja Burgard ◽  
Michael Bošnjak
Keyword(s):  
Structural Equation Modeling ◽  
Structural Equation ◽  
Factor Model ◽  
Self Report ◽  
Equation Modeling ◽  
Positive And Negative Affect ◽  
Correlation Matrices ◽  
The World ◽  
Best Fit ◽  
Cross National

Abstract. The Positive and Negative Affect Schedule (PANAS; Watson et al., 1988 ) is a popular self-report questionnaire that is administered all over the world. Though originally developed to measure two independent factors, different models have been proposed in the literature. Comparisons among alternative models as well as analyses concerning their robustness in cross-national research have left an inconclusive picture. Therefore, the present study evaluates the dimensionality of the PANAS and differences between English and translated versions of the PANAS using a meta-analytic structural equation modeling approach. Correlation matrices from 57 independent samples ( N = 54,043) were pooled across subsamples. For both English and non-English samples, a correlated two-factor model including correlated uniquenesses provided the best fit. However, measurement invariance analyses indicated differences in factor loadings between subsamples. Thus, cross-national application of the PANAS might only be justified if measurement equivalence was explicitly tested for the countries at hand.

scholarly journals Explanation of Children’s Health through Parents’ Mental Health and Job Satisfaction by Structural Equation Modeling

2016 ◽  
Vol 9 (2) ◽  
pp. 166
Author(s):  
Majid Golzarpour ◽  
Meroe Vameghi ◽  
Homeira Sajjadi ◽  
Gholamreza Ghaedamini Harouni
Keyword(s):  
Mental Health ◽  
Job Satisfaction ◽  
Structural Equation Modeling ◽  
Structural Equation ◽  
Goodness Of Fit ◽  
General Health Questionnaire ◽  
Children's Health ◽  
Children’S Health ◽  
Equation Modeling ◽  
Health Questionnaire

<p><strong>BACKGROUND:</strong> Worldwide, much evidence exists on the influence of parents’ socioeconomic conditions, including employment, on children’s health. However, the mechanisms for this affect are still being investigated. Few studies have been conducted in Iran to investigate this issue. This study investigated working conditions, job satisfaction, and mental health of employed people and the association between these variables and their children’s health.<strong></strong></p><p><strong>MATERIALS &amp; METHODS:</strong> In this correlational work, 200 male and female staff of the official part of Educational Organization and the schools of Mashhad with children aged 5-18 years was randomly selected. The data were gathered using a demographic questionnaire, the 20-item Minnesota Job Satisfaction Questionnaire, the 28-item General Health Questionnaire, and the 28-item Child Health Questionnaire. The data were then analyzed using SPSS. The associations under study were investigated by structural equation modeling in AMOS.<strong></strong></p><p><strong>RESULTS:</strong> Approximately 17% of the variation in the parents’ job satisfaction could be explained by the parents’ insurance, income, and work hours; 6% of the variation in their mental health was explained by job satisfaction, and 26% of the variation in children’s health was directly explained by the parents’ job satisfaction and mental health. However, approximately 32.2% of the variation in children’s health could be explained in the light of the direct effect of the parents’ mental health and direct and indirect effects of the parents’ job satisfaction. The goodness of fit index was 0.94.</p><p><strong>CONCLUSION:</strong> Parents’ job satisfaction was associated with and considerably explained children’s health. Although this finding may be partially related to the job satisfaction effect on mental health, the reasons for the affect of job satisfaction on children’s health and the potential mechanisms of this association require further studies.<strong></strong></p>

scholarly journals MODEL PENGUKURAN TINGKAT KEPERCAYAAN PELANGGAN (ONLINE TRUST) TERHADAP SITUS E-COMMERCE (Studi Kasus pada Pelanggan E-Commerce di Provinsi Bali)

2018 ◽  
Vol 1 (3) ◽  
pp. 100
Author(s):  
I Made Endra Wiartika Putra ◽  
Gede Rasben Dantes ◽  
I Made Candiasa
Keyword(s):  
Structural Equation Modeling ◽  
Structural Equation ◽  
Goodness Of Fit ◽  
Equation Modeling ◽  
Snowball Sampling ◽  
Cronbach’S Alpha ◽  
Purposive Sampling ◽  
Online Trust ◽  
Cronbach's Alpha ◽  
Fit Index

Penelitian ini bertujuan untuk mengetahui model pengukuran tingkat kepercayaan pelanggan terhadap situs e-commerce. Langkah awal yang dilakukan yaitu identifikasi faktor-faktor yang mempengaruhi kepercayaan pelanggan melalui studi literatur dan studi empirik untuk menentukan model analisis terhadap kepuasan pelanggan. Faktor yang mempengaruhi kepercayaan pelanggan untuk bertransaksi secara online yaitu pengetahuan konsumen terhadap e-commerce, reputasi penjual, resiko dalam transaksi, kemudahan penggunaan e-commerce, jaminan sistem, sikap/perilaku terhadap sistem dan sistem keamanan. Populasi dalam penelitian ini adalah masyarakat Provinsi Bali menggunakan metode purposive sampling dan snowball sampling dengan kriteria responden pernah berkunjung dan melakukan transaksi di e-commerce yang ada di Indonesia lebih dari 3 kali. Instrumen penelitian berupa kuesioner dengan data interval berskala 5 Likert. Instrumen terlebih dahulu diuji validitas isi dengan metode Robert Gregory, validitas empiris menggunakan rumus product moment, reliabilitas instrument menggunakan Cronbach’s Alpha, dan menghasilkan 59 pernyataan yang dapat digunakan untuk pengambilan data. Jumlah responden yang digunakan dalam penelitian ini adalah sebanyak 126 responden. Teknik analisis data, pengujian hipotesis dan pengujian model menggunakan metode Structural Equation Modeling dengan bantuan aplikasi SPSS AMOS 21. Hasil penelitian ini melalui pengujian hipotesis menunjukkan bahwa pengetahuan tentang situs e-commerce dan perlindungan keamanan berpengaruh negatif dan tidak signifikan terhadap kepercayaan pelanggan. Resiko, kemudahan e-commerce, jaminan sistem dan sistem keamanan bukan menjadi sesuatu yang penting untuk dipertimbangkan dalam meningkatkan kepercayaan pelanggan karena pengaruhnya tidak signifikan. Reputasi yang dirasakan dan sikap merupakan hal yang perlu diperhatikan dan paling berpengaruh terhadap kepercayaan pelanggan pelanggan. Hasil penelitian ini kemudian diuji menggunakan goodness of fit index dan menghasilkan bahwa model penelitian tersebut dapat diterima dan dapat digunakan untuk meningkatkan keinginan pelanggan untuk bertransaksi online

scholarly journals Cutoff criteria for overall model fit indexes in generalized structured component analysis

2020 ◽  
Vol 8 (4) ◽  
pp. 189-202
Author(s):  
Gyeongcheol Cho ◽  
Heungsun Hwang ◽  
Marko Sarstedt ◽  
Christian M. Ringle
Keyword(s):  
Structural Equation Modeling ◽  
Structural Equation ◽  
Goodness Of Fit ◽  
Component Analysis ◽  
Model Fit ◽  
Equation Modeling ◽  
Generalized Structured Component Analysis ◽  
Root Mean Square Residual ◽  
Fit Indexes ◽  
Cutoff Criteria

AbstractGeneralized structured component analysis (GSCA) is a technically well-established approach to component-based structural equation modeling that allows for specifying and examining the relationships between observed variables and components thereof. GSCA provides overall fit indexes for model evaluation, including the goodness-of-fit index (GFI) and the standardized root mean square residual (SRMR). While these indexes have a solid standing in factor-based structural equation modeling, nothing is known about their performance in GSCA. Addressing this limitation, we present a simulation study’s results, which confirm that both GFI and SRMR indexes distinguish effectively between correct and misspecified models. Based on our findings, we propose rules-of-thumb cutoff criteria for each index in different sample sizes, which researchers could use to assess model fit in practice.

Organizational Commitment As Influenced by Leadership Behavior of Administrators, Job Satisfaction and Burnout of Teachers: A Structural Equation Modeling

10.17158/228 ◽  
2012 ◽  
Vol 18 (1) ◽  
Author(s):  
Felix C. Chavez, Jr.
Keyword(s):  
Job Satisfaction ◽  
Structural Equation Modeling ◽  
Organizational Commitment ◽  
Structural Equation ◽  
Quantitative Research ◽  
Goodness Of Fit ◽  
Leadership Behavior ◽  
Stepwise Multiple Regression ◽  
Equation Modeling ◽  
Best Fit

This study was conducted to determine the best fit model of organizational commitment. Specifically, it established the interrelationship among leadership behavior, job satisfaction, burnout, and organizational commitment. Quantitative research design was utilized in this study. The data were gathered from the teachers among the randomly selected academic institutions in Region XI, Philippines. Moreover, sets of survey questionnaires were used as instruments to obtain information from the participants. Pearson product moment correlation was used to find the significance of the relationship between the independent and dependent variables. Stepwise multiple regression analysis was used to identify the variables that best predict organizational commitment and likewise Structural Equation Modeling was used to identify the model that best fits organizational commitment. The findings revealed that the over-all leadership behavior of administrators and organizational commitment of teachers were high. On the other hand, the job satisfaction of teachers was moderate and their degree of burnout was low. Furthermore, the leadership behavior, job satisfaction, and burnout were highly correlated with organizational commitment, and found to be significant predictors of organizational commitment. Finally, the best fit model of organizational commitment was the Hypothesized Model 5, which passed all the goodness of fit indices criteria.

Sign in / Sign up

Export Citation Format

Share Document