Message-ID: <39B6DDB9048D0F4DAD42CB26AAFF0AFA070235F6@usctmx1106.merck.com>
Date: 2009-03-24T14:55:02Z
From: Liaw, Andy
Subject: Do we have to control for block in block designs if it is insignificant?
In-Reply-To: <BAY134-W7F1495AB642F1B471DABF88930@phx.gbl>
The short answer is "no" (meaning to leave the blocks in the model). As
Frank Harrell said, you've spent your degrees of freedom. Go home and
be happy.
Best,
Andy
From: J S
> Sent: Tuesday, March 24, 2009 9:49 AM
> To: r-help at r-project.org
> Subject: [R] Do we have to control for block in block designs
> if it is insignificant?
>
>
>
>
>
> I am wondering if in a block experimental design (ex. triple
> square lattice), the block effect was not significant, is it
> all right not to include the block effect in an empirical
> model (even though the sampling was done from different
> blocks)? Or we are forced to control for the block effect in
> block designs anyway?
>
> Thanks,
> Julia
>
>
> _________________________________________________________________
> Internet Explorer 8 - Now Available. Faster, safer, easier.
>
> [[alternative HTML version deleted]]
>
>
Notice: This e-mail message, together with any attachme...{{dropped:12}}