MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V. A comparison of methods to test mediation and other intervening variable effects

Department of Psychology, Arizona State University, Tempe 85287-1104, USA.
Psychological Methods (Impact Factor: 4.45). 04/2002; 7(1):83-104. DOI: 10.1037/1082-989X.7.1.83
Source: PubMed


A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.

Download full-text


Available from: David MacKinnon, Oct 10, 2015
103 Reads
    • "Thus, we estimated the indirect, direct, and total effects of brand personality dimensions upon visit intention. Furthermore, the T statistics for these effects are calculated using a Sobel test (MacKinnon et al., 2002 "
    [Show abstract] [Hide abstract]
    ABSTRACT: Tourism destinations increasingly use brand-personification strategies to evoke favorable consumer reactions. These reactions, however, may hinge on cultural differences. This paper investigates the relationships among nation brand personality perceptions, consumer brand-self congruity, and the visit intention of a country as a tourism destination. Brand-self congruity is examined as a mediator of the relationship between brand personality perception and visit intention. Of Hofstede's cultural dimensions, individualism and uncertainty avoidance are the most relevant dimensions for brand-self congruity. Based on representative samples of consumers from five countries (Italy, the UK, Czech Republic, Poland, and Russia) and using Slovakia as a sample tourism destination, the effect of individualism and uncertainty avoidance on the relationship between brand self-congruity and visit intention is studied. Individualism and uncertainty avoidance moderate the congruity e visit intention relationship e but in a negative way, contrary to our expectations. Important implications are derived for both tourism research and destination management.
    Tourism Management 02/2016; 52:507-520. DOI:10.1016/j.tourman.2015.07.017 · 2.57 Impact Factor
  • Source
    • "A closer inspection of the final equation in the second step of the analysis revealed that only anger rumination was a significant predictor of aggressive proneness, over and above emotional dysregulation. The subsequent normal theory test (MacKinnon et al., 2002) was statistically significant (z = 2.46; p = 0.01) with indirect effect bootstrap estimate = 0.11; 95% confidence interval [0.04, 0.22], supporting a significant mediating role of anger rumination in the relationship between emotion dysregulation and aggressive proneness, across the entire PD sample. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Emotional instability and dyscontrolled behaviours are central features in borderline personality disorder (BPD). Recently, some cognitive dysfunctional mechanisms, such as anger rumination, have been found to increase negative emotions and promote dyscontrolled behaviours. Even though rumination has consistently been linked to BPD traits in non-clinical samples, its relationship with problematic behaviour has yet to be established in a clinical population.AimThe purpose of the study was to explore the relationships between emotional dysregulation, anger rumination and aggression proneness in a clinical sample of patients with BPD.Methods Enrolled patients with personality disorders (93 with BPD) completed a comprehensive assessment for personality disorder symptoms, anger rumination, emotional dysregulation and aggression proneness.ResultsAnger rumination was found to significantly predict aggression proneness, over and above emotional dysregulation. Furthermore, both BPD diagnosis and anger rumination were significant predictors of aggression proneness.Conclusion Future research should examine whether clinical techniques aimed at reducing rumination are helpful for reducing aggressive and other dyscontrolled behaviours in treating patients with BPD. Copyright © 2015 John Wiley & Sons, Ltd.
    Personality and Mental Health 09/2015; DOI:10.1002/pmh.1310 · 1.10 Impact Factor
  • Source
    • "There were many combinations of the four parameter values. We chose the following values because they cover the range of values in past simulation studies (Taylor et al., 2008; MacKinnon et al., 2002) and make the simulation study more manageable. To study the Type I error rate for b 1 b 2 b 3 b 4 , we set b 4 = 0 and b 1 = b 2 = b 3 = 0, 0.14, 0.39 and 0.59. "
    [Show abstract] [Hide abstract]
    ABSTRACT: One challenge in mediation analysis is to generate a confidence interval (CI) with high cov- erage and power that maintains a nominal significance level for any well-defined function of indirect and direct effects in the general context of structural equation modeling (SEM). This study discusses a proposed Monte Carlo extension that finds the CIs for any well-defined function of the coefficients of SEM such as the product of k coefficients and the ratio of the contrasts of indirect effects, using the Monte Carlo method. Finally, we conduct a small-scale simulation study to compare CIs produced by the Monte Carlo, nonparametric bootstrap, and asymptotic-delta methods. Based on our simulation study, we recommend researchers use the Monte Carlo method to test a complex function of indirect effects.
    Structural Equation Modeling A Multidisciplinary Journal 08/2015; DOI:10.1080/10705511.2015.1057284 · 4.18 Impact Factor
Show more