[R] model selection for nested factorial design

Ben Bolker bbolker at gmail.com
Wed May 14 23:04:07 CEST 2014


K C <interlocutorbrl2 <at> gmail.com> writes:

> 

  [snip]

> I have a dataset composed of observations taken from 16 separate
> experimental panels, each nested into one of 4 conditions (Treatment A
> Level 1, Treatment A Level 2, Treatment B Level 1, Treatment 
> B Level 2; see
> photo: http://imgur.com/ZbzFPNq). There are 100 observations of the
> dependent variable for each of the 16 panels (1600 total obs).
> 
> I'm trying to determine main effects of both treatment types, and any
> interaction effect, accounting for the within-panel variation. I'm trying
> to determine if the DV differs significantly across conditions.

    This is a stats question, not an R question per se ... I would
recommend posting to CrossValidated (http://stats.stackexchange.com)

  good luck,
     Ben Bolker



More information about the R-help mailing list