Principle of marginality

From Wikipedia, the free encyclopedia
Jump to: navigation, search

In statistics, the principle of marginality refers to the fact that the average (or main) effects, of variables in an analysis are marginal to their interaction effect.[clarification needed] The principle of marginality argues that, in general, it is wrong to test, estimate, or interpret main effects of explanatory variables where the variables interact or, similarly, to model interaction effects but delete main effects that are marginal to them.[1] While such models are interpretable, they lack applicability, as they ignore the effects of their marginal main effects.

Nelder[2] and Venables[3] have argued strongly for the importance of this principle in regression analysis.

See also[edit]


  1. ^ Fox, J. Regression Notes.
  2. ^ Nelder, J. A. (1977). "A Reformulation of Linear Models". Journal of the Royal Statistical Society. 140 (1): 48–77. doi:10.2307/2344517. 
  3. ^ Venables, W.N. (1998). "Exegeses on Linear Models". Paper presented to the S-PLUS User’s Conference Washington, DC, 8–9 October 1998.

External links[edit]