Methods of Meta-Analysis: Correcting Error and Bias in Research Findings

Hunter, John E.; Schmidt, Frank L.

doi:10.4135/9781483398105

Cited by 2,133 publications

(2,200 citation statements)

References 0 publications

Supporting

Mentioning

2,135

Contrasting

Unclassified

Order By: Relevance

“…A meta-analysis aims to obtain a set of objective, replicable and accurate statistical data [170] that provide additional evidence that is drawn from the entire sample of the studies analyzed, and that it could not be obtained from individual studies [171,172]. Two main statistical models have been applied in previous meta-analyses: (i) the fixed effects model; and, (ii) the random effects model [173,174].…”

Section: Meta-analytic Proceduresmentioning

confidence: 99%

“…To analyze the homogeneity of the observed correlations, the Cochram's Q statistic [172,177] is computed by Equation (5):…”

Section: Meta-analytic Proceduresmentioning

confidence: 99%

“…For those papers providing various effect sizes (i.e., reporting two correlation coefficients between the independence of a firm's board and environmental performance and social performance respectively), we followed the approach adopted by Hunter and Schmidt [172] of computing the average correlation [67]. Articles included in the final sample were coded by addressing the following issues: (i) authors; (ii) year of publication; (iii) CSP measurement model; (iv) correlation coefficient (observed or calculated); (v) countries covered by the sample; (vi) CG systems covered by the sample; and, (vii) sample period (see Table A1 in the Appendix A for detailed information).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Firms’ Board Independence and Corporate Social Performance: A Meta-Analysis

Ortas

Álvarez

Zubeltzu³

2017

Sustainability

View full text Add to dashboard Cite

This paper investigates the influence of organizations' board independence on corporate social performance (CSP) using a meta-analytic approach. A sample of 87 published papers is used to identify a set of underlying moderating effects in that relationship. Specifically, differences in the system of corporate governance, CSP measurement models and market conditions have been considered as moderating variables. The results show that the independence of a company's board positively influences CSP. This is because companies with more independent directors in their boards are more likely to commit to stakeholder engagement, environmental preservation and community well-being. Interestingly, the results also show that the positive connection between board independence and CSP is stronger in civil law countries and when CSP is measured by self-reporting data. Finally, the strength of the influence of the independence of a firm's board on CSP varies significantly in different market conditions. The paper concludes by presenting the main implications for academics, practitioners and policy makers.

show abstract

Section: Meta-analytic Proceduresmentioning

confidence: 99%

“…To analyze the homogeneity of the observed correlations, the Cochram's Q statistic [172,177] is computed by Equation (5):…”

Section: Meta-analytic Proceduresmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Firms’ Board Independence and Corporate Social Performance: A Meta-Analysis

Ortas

Álvarez

Zubeltzu³

2017

Sustainability

View full text Add to dashboard Cite

show abstract

“…Different solutions for dealing with dependency of effect sizes have been described in the literature (see, for instance, Borenstein et al, 2009;Cooper, 2010;Del Re, 2015;Hedges & Olkin, 1985;Lipsey & Wilson, 2001;Rosenthal, 1984;Schmidt & Hunter, 2015). Common methods for handling dependency of effect sizes are: simply ignoring the dependency and analyzing the effect sizes as if they were independent; averaging the dependent effect sizes within studies into a single effect size by calculating an unweighted or -less biased -weighted average; selecting only one effect size per study (also referred to as eliminating effect sizes); and shifting the unit of analysis meaning that one unit of analysis is selected after which effect sizes are averaged within each unit.…”

Section: Introductionmentioning

confidence: 99%

Fitting three-level meta-analytic models in R: A step-by-step tutorial

Assink¹,

Wibbelink²

2016

TQMP

790

846

View full text Add to dashboard Cite

Applying a multilevel approach to meta-analysis is a strong method for dealing with dependency of effect sizes. However, this method is relatively unknown among researchers and, to date, has not been widely used in meta-analytic research. Therefore, the purpose of this tutorial was to show how a three-level random effects model can be applied to meta-analytic models in R using the rma.mv function of the metafor package. This application is illustrated by taking the reader through a step-by-step guide to the multilevel analyses comprising the steps of (1) organizing a data file; (2) setting up the R environment; (3) calculating an overall effect; (4) examining heterogeneity of within-study variance and between-study variance; (5) performing categorical and continuous moderator analyses; and (6) examining a multiple moderator model. By example, the authors demonstrate how the multilevel approach can be applied to meta-analytically examining the association between mental health disorders of juveniles and juvenile offender recidivism. In our opinion, the rma.mv function of the metafor package provides an easy and flexible way of applying a multi-level structure to meta-analytic models in R. Further, the multilevel meta-analytic models can be easily extended so that the potential moderating influence of variables can be examined.

show abstract

“…A major task in all areas of science is the development of theory and theoretical concepts, ultimately the production of cumulative knowledge (Schmidt, 2014) and to model a phenomenon at a deeper level (Heckman, 2005). The pathways to it entail the finding of the objectively sustainable meaning of key notions and the subjective awareness of relevant groups in this respect.…”

Section: Sources and Methodsmentioning

confidence: 99%