Skip to main content
SpringerLink
Log in

Analysis and evaluation of moderator effects in regression models: state of art, alternatives and empirical example

  • Original Paper
  • Published:
Review of Managerial Science Aims and scope Submit manuscript

Abstract

The identification and analysis of moderator relationships regularly confronts the empirical research with statistical and methodical challenges. Which misinterpretations and false conclusions result from different methodical procedures for the identification of moderator effects shall be demonstrated by means of the present contribution. Thereby, the moderated regression analysis represents the most popular procedure in this context.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. Implying an interaction effect of the independent variables x and z on the dependent variable y, on the one hand, one can assume that the effect of x on y is moderated by z; on the other hand, the correlation can also be interpreted in a way that the effect of z on y is moderated by x. Insofar the variables x and z act symmetrically. Which variable is defined as moderator variable is an issue for theoretical consideration.

  2. In the context of interpreting first-order effects often the misleading term “main effect” is used, possibly following ANOVA wording,. Numerous authors refer to this vagueness of conception (Baltes-Götz 2006; Einhorn and Bass 1971; Hunter and Schmidt 1978).

  3. On the supposition that x and y act symmetrically for reasons of completeness, also Model III is displayed; see also the annotations in footnote 2. Model III implies that z serves as predictor and x as moderator variable.

References

  • Aguinis H (1995) Statistical power problems with moderated multiple regression in management research. J Manag 21:1141–1158

    Google Scholar 

  • Aguinis H (2004) Regression analysis for categorical moderators. Guilford Press, New York

    Google Scholar 

  • Aguinis H, Beaty JC, Boik RJ, Pierce CA (2005) Effect size and power in assessing moderating effects of categorical variables using multiple regression: a 30- year review. J Appl Psychol 90(1):94–107

    Article  Google Scholar 

  • Aiken LS, West SG (1991) Multiple regression: testing and interpreting interactions. Sage, Newbury Park, CA

    Google Scholar 

  • Anderson LE, Stone-Romero EF, Tisak JA (1996) A comparison of bias and mean squared error in parameter estimates of interaction effects: moderated multiple regression versus errors-in-variables regression. Multivar Behav Res 31:69–94

    Article  Google Scholar 

  • Arnold HJ (1982) Moderator variables: a clarification of conceptual, analytic, and psychometric issues. Organ Behav Hum Perform 29:143–174

    Article  Google Scholar 

  • Arnold HJ (1984) Testing moderator variable hypotheses: a reply to Stone and Hollenbeck. Organ Behav Hum Perform 34:214–224

    Article  Google Scholar 

  • Baltes-Götz B (2006) Moderatoranalyse per multipler Regression mit SPSS, Online-Dokument: http://www.uni-trier.de/urt/urthome.shtml, abgerufen am 15 July 2006

  • Baron RM, Kenny DA (1986) The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 51(6):1173–1182

    Article  Google Scholar 

  • Bissonnette V, Ickes W, Bernstein I, Knowles E (1990) Personality moderating variables: a warning about statistical artifact and a comparison of analytic techniques. J Pers 58:567–587

    Article  Google Scholar 

  • Carte TA, Russell CJ (2003) In pursuit of moderation: nine common errors and their solutions. MIS Q 27(3):479–501

    Google Scholar 

  • Chaplin WF (1991) The next generation in moderation research in personality psychology. J Pers 59:143–178

    Article  Google Scholar 

  • Cohen J (1988) Statistical power analysis for the behavioral sciences, 2nd edn. Erlbaum, Hillsdale, NJ

    Google Scholar 

  • Cohen J, Cohen P, West SG, Aiken LS (2003) Applied multiple regression/correlation analyses for the behavioral sciences, 3rd edn. Lawrence Erlbaum, Mahwah, NJ

    Google Scholar 

  • Cortina JM (1993a) Interaction, nonlinearity, and multicollinearity: implications for multiple regression. J Manag 19:915–922

    Google Scholar 

  • Cortina JM (1993b) What is coefficient alpha? an examination of theory and applications. J Appl Psychol 78:98–104

    Article  Google Scholar 

  • Cronbach LJ (1987) Statistical tests for moderator variables: flaws in analyses recently proposed. Psychol Bull 102:414–417

    Article  Google Scholar 

  • Darrow AL, Kahl DR (1982) A comparison of moderated regression techniques considering strength of effect. J Manag 8(2):35–47

    Article  Google Scholar 

  • Einhorn HJ, Bass AR (1971) Methodological considerations relevant to discrimination in employment testing. Psychol Bull 75:261–269

    Article  Google Scholar 

  • Frazier PA, Tix AP, Barron KE (2004) Testing moderator and mediator effects in counseling psychology research. J Couns Psychol 51(1):115–134

    Article  Google Scholar 

  • Hays WL (1983) Statistics. Holt Rinehart & Winston, New York

    Google Scholar 

  • Holmbeck GN (1997) Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: examples from the child-clinical and pediatric psychology literatures. J Consult Clin Psychol 65:599–610

    Article  Google Scholar 

  • Huber F, Heitmann M, Herrmann A (2006) Ansätze zur Kausalmodellierung mit Interaktionseffekten. Die Betriebswirtschaft 66(6):696–710

    Google Scholar 

  • Hunter JE, Schmidt FL (1978) Differential and single-group validity of employment tests by race: a critical analysis of three recent studies. J Appl Psychol 63:1–11

    Article  Google Scholar 

  • Irwin JR, McClelland GH (2001) Misleading heuristics and moderated multiple regression models. J Mark Res 38:100–109

    Article  Google Scholar 

  • Jaccard J, Turrisi R (2003) Interaction effects in multiple regression, 2nd edn. Sage, Newbury Park, CA

    Google Scholar 

  • Jaccard J, Wan CK (1995) Measurement error in the analysis of interaction effects between continuous predictors using multiple regression: multiple indicator and structural equation approaches. Psychol Bull 117:348–357

    Article  Google Scholar 

  • Jaccard J, Wan CK (1996) LISREL approaches to interaction effects in multiple regression. Thousand Oaks, London, New Delhi

  • Jaccard J, Turrisi R, Wan CK (1990) Interaction effects in multiple regression, Sage University paper series on quantitative applications in the social sciences 07–072. Sage, Newbury Park, CA

    Google Scholar 

  • James LR, Brett JM (1984) Mediators, moderators, and tests for mediation. J Appl Psychol 69:307–321

    Article  Google Scholar 

  • Judd CM, McClelland GH, Culhane SE (1995) Data analysis: continuing issues in the everyday analysis of psychological data. Annu Rev Psychol 46:433–465

    Article  Google Scholar 

  • Judd CM, Kenny DA, McClelland GH (2001) Estimating and testing mediation and moderation in within-subjects designs. Psychol Methods 6:115–134

    Article  Google Scholar 

  • Kahl DR, Darrow AL (1984) Model determination in moderated regression. J Manag 10:234–236

    Article  Google Scholar 

  • Landis RS, Dunlap WP (2000) Moderated multiple regression–tests are criterion specific. Organ Res Methods 3:254–266

    Article  Google Scholar 

  • Lubinski D, Humphreys LG (1990) Assessing spurious “moderator effects”: illustrated substantively with the hypothesized (“synergistic”) relation between spatial and mathematical ability. Psychol Bull 107:385–393

    Article  Google Scholar 

  • MacCallum RC, Mar CM (1995) Distinguishing between moderator and quadratic effects in multiple regression. Psychol Bull 118:405–421

    Article  Google Scholar 

  • MacCallum RC, Zhang S, Preacher KJ, Rucker DD (2002) On the practice of dichotomization of quantitative variables. Psychol Methods 7:19–40

    Article  Google Scholar 

  • Mason CA, Tu S, Cauce AM (1996) Assessing moderator variables: two computer simulation studies. Educ Psychol Measur 56:45–62

    Article  Google Scholar 

  • McClelland GH, Judd CM (1993) Statistical difficulties of detecting interactions and moderator effects. Psychol Bull 114:376–390

    Article  Google Scholar 

  • Morris JH, Mansfield ER, Sherman JD (1986) Failures to detect moderating effects with ordinary least squares-moderated multiple regression: some reasons and a remedy. Psychol Bull 99:282–288

    Article  Google Scholar 

  • Muller D, Judd CM, Yzerbyt VY (2005) When moderation is mediated and mediation is moderated. J Pers Soc Psychol 89(6):852–863

    Article  Google Scholar 

  • Nunnally JC (1967) Psychometric theory. McGraw-Hill, New York

    Google Scholar 

  • Nunnally JC (1978) Psychometric theory, 2nd edn. McGraw-Hill, New York

    Google Scholar 

  • Peterson RA (1994) A meta-analysis of cronbach’s coefficient alpha. J Consum Res 21:381–391

    Article  Google Scholar 

  • Schermelleh-Engel K, Moosbrugger H, Klein A (1998) Structural equation models with latent interaction effects: comparing the efficiency of LMS, LISREL-ML and LISREL-WLSA. In: Hox JJ, de Leeuw ED (eds) Assumptions, robustness, and estimation methods in multivariate modeling. TT-Publikaties, Amsterdam, pp 87–109

    Google Scholar 

  • Schmitt NW, Klimoski RJ (1991) Research methods in human resource management. South-Western, Cincinnati, OH

    Google Scholar 

  • Sharma S, Durand RM, Gur-Arie O (1981) Identification and analysis of moderator variables. J Mark Res 18:291–300

    Article  Google Scholar 

  • Shrout PE, Bolger N (2002) Mediation in experimental and nonexperimental studies: new procedures and recommendations. Psychol Methods 7:422–445

    Article  Google Scholar 

  • Smith KW, Sasaki MS (1979) Decreasing multicollinearity: a method for models with multiplicative functions. Sociol Methods Res 8:35–56

    Article  Google Scholar 

  • Stone EF (1988) Moderator variables in research: a review and analysis of conceptual and methodological issues. In: Ferris GR, Rowland KM (eds) Research in personnel and human resources management, vol 6. JAI Press, Greenwich, pp 191–229

    Google Scholar 

  • Stone EF, Hollenbeck JR (1984) Some issues associated with the use of moderated regression. Organ Behav Hum Perform 34:195–213

    Article  Google Scholar 

  • Stone EF, Hollenbeck JR (1989) Clarifying some controversial issues surrounding statistical procedures for detecting moderator variables: empirical evidence and related matters. J Appl Psychol 74:3–10

    Article  Google Scholar 

  • Stone-Romero EF, Anderson LE (1994) Relative power of moderated multiple regression and the comparison of subgroup correlation coefficients for detecting moderating effects. J Appl Psychol 79(3):354–359

    Article  Google Scholar 

  • Zedeck S (1971) Problems with the use of “moderator” variables. Psychol Bull 76:295–310

    Article  Google Scholar 

Download references

Acknowledgments

The authors would like to thank two anonymous reviewers and Wolfgang Kuersten for helpful comments on earlier versions of this paper.

Author information

Authors and Affiliations

  1. Chair of Strategic Industrial Marketing, University of Regensburg, 93053, Regensburg, Germany

    Roland Helm

  2. Polymet Jena e.V., 07745, Jena, Germany

    Antje Mark

Authors
  1. Roland Helm
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Antje Mark
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Roland Helm.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Helm, R., Mark, A. Analysis and evaluation of moderator effects in regression models: state of art, alternatives and empirical example. Rev Manag Sci 6, 307–332 (2012). https://doi.org/10.1007/s11846-010-0057-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11846-010-0057-y

Keywords

JEL Classification

Search

Navigation