A basic introduction to fixed‐effect and random‐effects models for meta‐analysis

Research Synthesis Methods 11/2010; 1(2):97 - 111. DOI: 10.1002/jrsm.12

ABSTRACT There are two popular statistical models for meta-analysis, the fixed-effect model and the random-effects model. The fact that these two models employ similar sets of formulas to compute statistics, and sometimes yield similar estimates for the various parameters, may lead people to believe that the models are interchangeable. In fact, though, the models represent fundamentally different assumptions about the data. The selection of the appropriate model is important to ensure that the various statistics are estimated correctly. Additionally, and more fundamentally, the model serves to place the analysis in context. It provides a framework for the goals of the analysis as well as for the interpretation of the statistics.In this paper we explain the key assumptions of each model, and then outline the differences between the models. We conclude with a discussion of factors to consider when choosing between the two models. Copyright © 2010 John Wiley & Sons, Ltd.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Whether high or low self-esteem is associated with increased aggression remains a topic of debate. The majority have argued that aggression is linked to low self-esteem, though this stance has been disputed by others. The present study aimed to explore the relationship between aggression and self-esteem by meta-analysis to provide conclusive evidence on this debate. Fifty-two studies were included, with a total of 82,358 Chinese student participants. The results suggested a medium negative correlation between aggression and self-esteem (r = -.21, 95% confidence interval [-.23, -.17]). Analysis of aggression sub-factors showed that almost all subtypes, except verbal aggression (i.e., physical aggression, anger, hostility, and implicit and explicit aggression), were negatively correlated with self-esteem. Moderator analyses suggested that various study and participant characteristics (i.e., subject group, sample size, and aggression assessment instrument) influenced the strength of the association between self-esteem and aggression.
    Aggression and Violent Behavior 01/2015; · 1.95 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A diagnosis of sadism in sexual offenders is commonly regarded as indicative of high risk for violent reoffending. The purpose of the current two studies was to evaluate whether sadism is indeed associated with higher rates of violent (including sexual) reoffending. In Study 1 (meta-analysis), the rate of violent and sexual recidivism was assessed across seven samples of male sex offenders (total N = 2,169) as a function of diagnoses of sexual sadism. In Study 2 (N = 768) the outcome (violent recidivism yes/no) was regressed on sadism, along with behavioral indicators of sexually sadistic offending, and scores from violence risk assessment instruments. In Study 1 (meta-analysis), the overall risk of sadists compared with nonsadists with respect to violent (including sexual contact) reoffending was slightly elevated (by a factor of 1.18), yet not significantly increased. Similarly, the risk of sexual reoffending among sadists was slightly, but not significantly, higher than among nonsadists (factor 1.38). According to Study 2, only a measure of sadistic behavior, not the clinical diagnosis, was associated with violent reoffending. This association, however, was not present once age and customary risk assessment instruments for violence risk were included in the regression. A clinical diagnosis of sexual sadism and behavioral measures of sadism are related to the risk of violent reoffending in sexual offenders. These associations, however, are weak and do not hold once variables relevant for the prediction of violence are controlled for. At the individual level, the risk for future violence in sadists can therefore be adequately described by customary risk assessment instruments. © The Author(s) 2015.
    Sexual Abuse A Journal of Research and Treatment 01/2015; · 1.54 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine the relationship between CD28 polymorphisms, rs3116496, and cancer. Meta-analysis. PubMed, EMBASE, Web of Science, and Cochrane library databases were searched to identify studies reporting the association between CD28 polymorphism and cancer. Two authors selected identified studies, extracted, and analyzed the data independently. Individuals carrying a T allele (TT homozygotes and TT+TC heterozygotes) at rs3116496 had a lower incidence of cancer than carriers of a C allele. Subgroup analysis showed that this association held true for Asians, but not Europeans. CD28 polymorphism, rs3116496, contributes to cancer susceptibility in the case of multiple cancers. Copyright © 2014. Published by Elsevier Ltd.
    Computers in Biology and Medicine 12/2014; · 1.48 Impact Factor


Available from
May 21, 2014