A General Model for Testing Mediation and Moderation Effects

Department of Psychology, University of South Carolina, Barnwell College, 1512 Pendleton St., Columbia, SC 29208, USA.
Prevention Science (Impact Factor: 2.63). 12/2008; 10(2):87-99. DOI: 10.1007/s11121-008-0109-6
Source: PubMed

ABSTRACT This paper describes methods for testing mediation and moderation effects in a dataset, both together and separately. Investigations of this kind are especially valuable in prevention research to obtain information on the process by which a program achieves its effects and whether the program is effective for subgroups of individuals. A general model that simultaneously estimates mediation and moderation effects is presented, and the utility of combining the effects into a single model is described. Possible effects of interest in the model are explained, as are statistical methods to assess these effects. The methods are further illustrated in a hypothetical prevention program example.

Download full-text


Available from: David MacKinnon, Sep 28, 2015
56 Reads
  • Source
    • "In other to improve the efficiency of MMR technique, Anderson et al (1996) recommended the use of small sample size Cortina (1993) recommended the use of square terms as covariates. Fairchild and MacKinnon (2009) wrote on a general model that has the capability of estimating both mediation and moderation effects simultaneously. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The paper introduces the concept of reverse moderation in order to investigate the uniqueness of the coefficients of independent variables and non-commutative nature of interactions in moderated multiple regression (MMR) in hierarchical order. The moderation effect is 0.01and the data used was masked to maintain the integrity of an ongoing research. The research concludes that moderation and its reverse yield different results indicating the uniqueness of the coefficients of the independent variables and the interactions are not commutative. Interactions are one-way. Each case is different as shown by the results of the 20 models used.
    • "Latent growth models to analyse change in intervention and control groups (Pentz and Chou, 1994), latent-class growth mixture models to examine the subsets that follow different change trajectories (Duncan and Duncan, 2004) and advanced techniques to estimate mediation models to examine the mechanisms that explain intervention-related change (Muthén et al., 2002) are now well-developed. There have been a number of applications of these methods to different fields of prevention in recent years (Cheong et al., 2003; Fairchild and Mackinnon, 2009) and modern statistical methodologies have made straightforward procedures available to test these models. Nevertheless, these advanced techniques are still underutilised in prevention research. "
    Journal of Children's Services 06/2015; 10(2):120-132. DOI:10.1108/JCS-02-2014-0016
  • Source
    • "The product of coefficients method for testing statistical mediation was applied using MPlus Version 5.2 (Muthén & Muthén, Los Angeles, LA), with percentile bootstrapping implemented to adjust asymmetric confidence limits and address biased standard errors (Fairchild, Mackinnon, Taborga, & Taylor, 2009; MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002). This method provides a balance of power and Type I error and supports the use of mediation when there may not be strong predictoreoutcome associations , whereas the causal steps and difference in coefficients methods are less advisable for relatively smaller samples, and are more susceptible to Type II errors (Fairchild & MacKinnon, 2009; Fritz & MacKinnon, 2007; MacKinnon et al., 2002). The product of coefficients method involves regression of outcomes on the mediator and predictors, and regression of the mediator on the predictors , yielding two coefficients that link predictors to the mediator and the mediator to the outcome, with the product of these coefficients providing an estimate of the mediated/indirect effect (ab). "
Show more