Accounting for Missing Correlation Coefficients in Fixed-Effects MASEM

ABSTRACT Meta-analytic structural equation modeling (MASEM) is increasingly applied to advance theories by synthesizing existing findings. MASEM essentially consists of two stages. In Stage 1, a pooled correlation matrix is estimated based on the reported correlation coefficients in the individual studies. In Stage 2, a structural model (such as a path model) is fitted to explain the pooled correlations. Frequently, the individual studies do not provide all the correlation coefficients between the research variables. In this study, we modify the currently optimal MASEM-method to deal with missing correlation coefficients, and compare its performance with existing methods. This study is the first to evaluate the performance of fixed-effects MASEM methods under different levels of missing correlation coefficients. We found that the often used univariate methods performed very poorly, while the multivariate methods performed well overall.

[1]  Olaf G. Rughase.  What is Organizational Identity , 2006 .

[2]  L. Hedges,et al.  Fixed- and random-effects models in meta-analysis. , 1998 .

[3]  A. Hafdahl.  Combining Correlation Matrices: Simulation Analysis of Improved Fixed-Effects Methods , 2007 .

[4]  Robbie C. M. van Aert,et al.  Meta-analysis using effect size distributions of only statistically significant studies. , 2015, Psychological methods.

[5]  A. Hafdahl,et al.  Combining Heterogeneous Correlation Matrices: Simulation Analysis of Fixed-Effects Methods , 2008 .

[6]  Robert Cudeck,et al.  Analysis of correlation matrices using covariance structure models. , 1989 .

[7]  A. Field.  Meta-analysis of correlation coefficients: a Monte Carlo comparison of fixed- and random-effects methods. , 2001, Psychological methods.

[8]  Wenmo Kong,et al.  Analyzing matrices of meta‐analytic correlations: current practices and recommendations , 2016, Research synthesis methods.

[9]  Betsy Jane Becker,et al.  Using Results From Replicated Studies to Estimate Linear Models , 1992 .

[10]  Leif D. Nelson,et al.  p-Curve and Effect Size , 2014, Perspectives on psychological science : a journal of the Association for Psychological Science.

[11]  R. Cooke,et al.  How well does the theory of planned behaviour predict alcohol consumption? A systematic review and meta-analysis , 2014, Health psychology review.

[12]  M. Hagger,et al.  Theory of planned behavior and adherence in chronic illness: a meta-analysis , 2015, Journal of Behavioral Medicine.

[13]  Mike W.-L. Cheung,et al.  metaSEM: an R package for meta-analysis using structural equation modeling , 2015, Front. Psychol..

[14]  D. Ones,et al.  Theory testing: Combining psychometric meta-analysis and structural equations modeling , 1995 .

[15]  Suzanne Jak,et al.  Meta-Analytic Structural Equation Modelling , 2015 .

[16]  Gabriela Topa,et al.  Theory of planned behavior and smoking: meta-analysis and SEM model , 2010, Substance Abuse and Rehabilitation.

[17]  D. Ketchen,et al.  Using meta-analytic structural equation modeling to advance strategic management research: Guidelines and an empirical illustration via the strategic leadership-performance relationship , 2016 .

[18]  Mike W.-L. Cheung,et al.  Testing IB theories with meta-analytic structural equation modeling , 2016 .

[19]  Mike W.-L. Cheung,et al.  A Two-Stage Approach to Synthesizing Covariance Matrices in Meta-Analytic Structural Equation Modeling , 2009 .

[20]  Leslie A. DeChurch,et al.  How Organizational Identity Affects Team Functioning: The Identity Instrumentality Hypothesis , 2018 .

[21]  John E. Hunter,et al.  Methods of Meta-Analysis: Correcting Error and Bias in Research Findings , 1991 .

[22]  Craig K. Enders,et al.  The Relative Performance of Full Information Maximum Likelihood Estimation for Missing Data in Structural Equation Models , 2001 .

[23]  Wolfgang Viechtbauer,et al.  Publication bias in meta-analysis: Prevention, assessment and adjustments , 2007, Psychometrika.

[24]  Mike W. L. Cheung,et al.  Comparison of methods for constructing confidence intervals of standardized indirect effects , 2009, Behavior research methods.

[25]  Adam R Hafdahl,et al.  Special issue on meta‐analytic structural equation modeling: introduction from the guest editors , 2016, Research synthesis methods.

[26]  Adam R Hafdahl,et al.  Meta-analysis of correlations revisited: attempted replication and extension of Field's (2001) simulation studies. , 2009, Psychological methods.

[27]  Victoria Savalei,et al.  On Obtaining Estimates of the Fraction of Missing Information From Full Information Maximum Likelihood , 2012 .

[28]  S Natasha Beretvas,et al.  Meta-analytic methods of pooling correlation matrices for structural equation modeling under different patterns of missing data. , 2005, Psychological methods.

[29]  A. Boomsma,et al.  Robustness Studies in Covariance Structure Modeling , 1998 .

[30]  S. Natasha Beretvas,et al.  An Evaluation of Multiple Imputation for Meta-Analytic Structural Equation Modeling , 2010 .

[31]  G. Loewenstein,et al.  Measuring the Prevalence of Questionable Research Practices With Incentives for Truth Telling , 2012, Psychological science.

[32]  P. Krishnaiah,et al.  On Covariance Structures. , 1975 .

[33]  Mike W-L Cheung,et al.  Meta-analytic structural equation modeling: a two-stage approach. , 2005, Psychological methods.

[34]  J. Vandekerckhove,et al.  A Bayesian approach to mitigation of publication bias , 2016, Psychonomic bulletin & review.

[35]  Betsy Jane Becker,et al.  Corrections to “Using Results From Replicated Studies to Estimate Linear Models” , 1995 .

[36]  Richard D Riley,et al.  Multivariate meta‐analysis: the effect of ignoring within‐study correlation , 2009 .

[37]  Peter M. Bentler,et al.  Covariance Structures Under Polynomial Constraints: Applications to Correlation and Alpha-Type Structural Models , 1983 .

[38]  M. Cheung,et al.  Applications of meta-analytic structural equation modelling in health psychology: examples, issues, and recommendations , 2017, Health psychology review.

[39]  Patrick J. Rosopa,et al.  Robustness of statistical inferences using linear models with meta-analytic correlation matrices , 2017 .

[40]  Suzanne Jak,et al.  Meta-analytic structural equation modelling with missing correlations , 2013 .

[41]  Reijo Sund,et al.  Meta‐Analysis: A Structural Equation Modeling Approach , 2016 .

[42]  I. Ajzen.  The theory of planned behavior , 1991 .

[43]  Leif D. Nelson,et al.  P-Curve and Effect Size: Correcting for Publication Bias Using Only Significant Results , 2014 .

[44]  Brian S. Connelly,et al.  A Meta-Analytic Multitrait Multirater Separation of Substance and Style in Social Desirability Scales. , 2016, Journal of personality.

[45]  L. Hedges,et al.  Statistical Methods for Meta-Analysis , 1987 .