Advanced search
1 file | 2.02 MB Add to list

Centering of interactions in lower-level mediation models

Author
Organization
Abstract
When considering multilevel mediation, centring is often applied to lower-level variables. One well-established approach for this separates the lower-level variables into a W(ithin)- and B(etween)-cluster component in linear settings, as to effectively eliminate additive upper level confounding of the mediator M and the outcome Y. When moderated mediation is considered, however, careful thought is needed about the method of centring; partitioning the interaction can be achieved in two ways: multiply the main effects that make up the interaction first, and apply centring within clusters next, or the other way around. Alternatively, M and Y can also be modelled jointly, hereby also allowing for unmeasured additive upper M-Y confounding, but at the same time avoiding any necessity for centring of both the main and the interaction effects. Employing simulations, we study the performance of these three approaches in the presence of interactions under varying data generating mechanisms, and discuss the relative merits of each approach.
Keywords
multilevel data, moderation, mediation analysis, centring

Downloads

  • JSM1.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 2.02 MB

Citation

Please use this url to cite or link to this publication:

MLA
Josephy, Haeike, and Tom Loeys. “Centering of Interactions in Lower-Level Mediation Models.” Joint Statistical Meetings 2016, Abstracts, 2016.
APA
Josephy, H., & Loeys, T. (2016). Centering of interactions in lower-level mediation models. In Joint Statistical Meetings 2016, Abstracts. Chicago, Il., U.S.A.
Chicago author-date
Josephy, Haeike, and Tom Loeys. 2016. “Centering of Interactions in Lower-Level Mediation Models.” In Joint Statistical Meetings 2016, Abstracts.
Chicago author-date (all authors)
Josephy, Haeike, and Tom Loeys. 2016. “Centering of Interactions in Lower-Level Mediation Models.” In Joint Statistical Meetings 2016, Abstracts.
Vancouver
1.
Josephy H, Loeys T. Centering of interactions in lower-level mediation models. In: Joint Statistical Meetings 2016, Abstracts. 2016.
IEEE
[1]
H. Josephy and T. Loeys, “Centering of interactions in lower-level mediation models,” in Joint Statistical Meetings 2016, Abstracts, Chicago, Il., U.S.A., 2016.
@inproceedings{8054531,
  abstract     = {{When considering multilevel mediation, centring is often applied to lower-level variables. One well-established approach for this separates the lower-level variables into a W(ithin)- and B(etween)-cluster component in linear settings, as to effectively eliminate additive upper level confounding of the mediator M and the outcome Y.  When moderated mediation is considered, however, careful thought is needed about the method of centring; partitioning the interaction can be achieved in two ways: multiply the main effects that make up the interaction first, and apply centring within clusters next, or the other way around. Alternatively, M and Y can also be modelled jointly, hereby also allowing for unmeasured additive upper M-Y confounding, but at the same time avoiding any necessity for centring of both the main and the interaction effects.  Employing simulations, we study the performance of these three approaches in the presence of interactions under varying data generating mechanisms, and discuss the relative merits of each approach.}},
  author       = {{Josephy, Haeike and Loeys, Tom}},
  booktitle    = {{Joint Statistical Meetings 2016, Abstracts}},
  keywords     = {{multilevel data,moderation,mediation analysis,centring}},
  language     = {{eng}},
  location     = {{Chicago, Il., U.S.A.}},
  pages        = {{20}},
  title        = {{Centering of interactions in lower-level mediation models}},
  year         = {{2016}},
}