Mean Centering in Moderated Multiple Regression: Much Ado about Nothing

Kromrey, Jeffrey D.; Foster‐Johnson, Lynn

doi:10.1177/0013164498058001005

Cited by 103 publications

(61 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although variable centering minimizes multicollinearity, its effectiveness is still disputable because centering does not affect any values of interest; that is, the regression coefficient, standard error, simple slope, t-test, and p-value of second order terms (the interaction term) are identical regardless of whether the continuous variable is centered (see Aiken & West, 1991;Kromrey & Foster-Johnson, 1998). Variable centering does not improve the power of MMR for detecting moderation effects but should be conducted with tests of mediated moderation so as to distinguish the impact of each variable and obtain meaningful regression results.…”

Section: Acknowledgmentsmentioning

confidence: 99%

The Mediated Moderation Model of Interactivity

2007

View full text Add to dashboard Cite

This paper argues for enhanced consideration of third variables in interactivity research and proposes a "mediated moderation" model to bring increased sophistication to bear on the study of information technology effects. Interactivity, a central phenomenon in new media research, is an elusive concept that has enduringly intrigued and confused scholars. Extant conceptualizations have produced incomplete causal models and have generally ignored the effect of third variables. We conceptualize interactivity as technological attributes of mediated environments that enable reciprocal communication or information exchange, which afford interaction between communication technology and users, or between users through technology. Specifying roles for mediator and moderator variables, this paper proposes a model that incorporates interactive attributes, user perceptions (mediators such as perceived interactivity), individual differences (moderators such as Internet self-efficacy), and media effects measures to systematically examine the definition, process, and consequences of interactivity on users. Lastly, statistical procedures for testing mediated moderation are described.An enduring question and major inconsistency in interactivity research is how to best isolate the concept for systematic investigation. Various definitions and multi-dimensional models have been proposed but current approaches attempt to either mix structural characteristics of media systems, message exchanges, and user perceptions into a single multidimensional construct, or identify one of these factors as the central locus of interactivity (Sundar, 2004). Not surprisingly, the empirical research on interactivity has yielded scattered findings and has been unable to ascertain consistent patterns of effects on users. After three decades of analysis and investigation, we scarcely know what interactivity really is, let alone what MEDIA PSYCHOLOGY, 9,

show abstract

Section: Acknowledgmentsmentioning

confidence: 99%

The Mediated Moderation Model of Interactivity

2007

View full text Add to dashboard Cite

show abstract

“…One increasingly common approach to the problem of multicollinearity in moderated multiple regression has been to center the constituent variables about their means before creating the interaction term (e.g., Aiken & West, 1991). As Kromrey and Foster-Johnson (1998) have shown, however, this approach does not change the interpretation or significance test of the interaction term. Moreover, in the case of correlated, continuous predictors with no meaningful 0 point, as we have here, tests of the main effects of the two predictor variables in the Joireman et al / COMMUTING PREFERENCES 203 …”

Section: Notesmentioning

confidence: 99%

Who Cares about the Environmental Impact of Cars?

Joireman

Lange

Vugt

2004

Environment and Behavior

245

201

View full text Add to dashboard Cite

This study examines preference for commuting to work by car or public transportation (PT) within an expanded social dilemma framework (i.e., one that recognizes the importance of both social and temporal concerns). Commuters completed scales assessing commuting preferences, beliefs regarding the environmental impact of cars, social value orientation (SVO), and the consideration of future consequences (CFC). Preference for PT was higher among commuters who believed that commut- ing by car harms the environment and among those scoring high in CFC. Most important, a significant two-way interaction revealed that preference for commuting by PT was positively related to beliefs regarding the harmful environmental consequences of commuting by car only among those high in CFC. SVO was unrelated to commuting preferences. In sum, a future orientation may be more important than a prosocial orientation in shaping commuting preferences.

show abstract

“…Researchers who do not believe the mean centering helps have no argument against mean centering per se; for example, if researchers are working with variables whose measurements include arbitrary zeros, then it may be fruitful to mean center a variable such that results are interpretable with respect to the variable's mean rather than to an arbitrary point of zero. Researchers of both camps mention the variables' measurement properties as a plausible and defensible reason for mean centering (Dalal & Zickar, 2012;Echambadi & Hess, 2007;Irwin & McClelland, 2001;Jaccard, Wan, & Turrisi, 1990;Kromrey & Foster-Johnson, 1998).…”

Section: Brief Literature Reviewmentioning

confidence: 99%

“…Allison, 1977;Dalal & Zickar, 2012;Kromrey & Foster-Johnson, 1998;Shieh, 2011Shieh, , 2010Shieh, , 2009Smith & Sasaki, 1979). Furthermore, most researchers concur that mean centering X 1 and X 2 will reduce their correlations with the product term X 1 X 2 .…”

Section: Brief Literature Reviewmentioning

confidence: 99%

“…Figure 2 shows the average over the 1,000 pairs of regressions for the β 1 (or β 2 ) estimates (given that the strength of the relationship between X 1 and Y was the same as that between X 2 and Y, both main effect βs were equal, for simplicity). As we mentioned previously, it is known that a linear transformation like mean centering will not affect the highest order term in a model, in this case, the X 1 X 2 interaction term (Allison, 1977;Dalal & Zickar, 2012;Kromrey & Foster-Johnson, 1998;Shieh, 2011Shieh, , 2010Shieh, , 2009Smith & Sasaki, 1979), thus we focus on the results for the main effect predictors. Much like the plot in Fig.…”

Section: Mean Centering Does Not Help Alleviate Bmacrom Ulticollinearitymentioning

confidence: 99%

See 1 more Smart Citation

Mean centering helps alleviate “micro” but not “macro” multicollinearity

et al. 2015

View full text Add to dashboard Cite

There seems to be confusion among researchers regarding whether it is good practice to center variables at their means prior to calculating a product term to estimate an interaction in a multiple regression model. Many researchers use mean centered variables because they believe it's the thing to do or because reviewers ask them to, without quite understanding why. Adding to the confusion is the fact that there is also a perspective in the literature that mean centering does not reduce multicollinearity. In this article, we clarify the issues and reconcile the discrepancy. We distinguish between Bmicro^and Bmacro^definitions of multicollinearity and show how both sides of such a debate can be correct. To do so, we use proofs, an illustrative dataset, and a Monte Carlo simulation to show the precise effects of mean centering on both individual correlation coefficients as well as overall model indices. We hope to contribute to the literature by clarifying the issues, reconciling the two perspectives, and quelling the current confusion regarding whether and how mean centering can be a useful practice.

show abstract

Mean Centering in Moderated Multiple Regression: Much Ado about Nothing

Cited by 103 publications

References 25 publications

The Mediated Moderation Model of Interactivity

The Mediated Moderation Model of Interactivity

Who Cares about the Environmental Impact of Cars?

Mean centering helps alleviate “micro” but not “macro” multicollinearity

Contact Info

Product

Resources

About