[R] To simplify or not simplify?

[Jeff Newmiller]

This forum is for questions about R. There are forums that focus on the theory of statistics (e.g. stats.stackexchage.com), but this particular issue is addressed in many statistics classes as well... and there is not necessarily a simple answer that always applies in all cases so be prepared to validate your model against a data set set aside for that purpose.
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
--------------------------------------------------------------------------- 
Sent from my phone. Please excuse my brevity.

On July 20, 2015 2:02:24 AM PDT, Georgina Southon <g.southon at sheffield.ac.uk> wrote:
>Dear R help,
>
>This is rather a basic question, but I can't seem to find an answer
>anywhere else.
>
>When I run a model such as lm/aov(height~var1) where var 1 is a
>categorical variable with 6 levels, I get output that shows some
>significant parameters and other non significant. Normally I would then
>proceed to simplify the model by removing the insignificant terms,
>however, I have recently begun to wonder if that should be standard
>practice or whether the full model output (not reduced by
>simplification) has more integrity and should be retained?
>
>Any thoughts would be most welcome!
>
>Thanks,
>
>Lizzie
>
>
>	[[alternative HTML version deleted]]
>
>______________________________________________
>R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide
>http://www.R-project.org/posting-guide.html
>and provide commented, minimal, self-contained, reproducible code.