Re: statistical question (somewhat obscure)

Rob Adelman (radelman@tvec.net)
Sat, 20 Jun 1998 19:52:46 -0500

I think the profiles of the patients with very tight construing i.e. 90% or
higher variance accounted by the first component may come close to what you
are thinking of. I had four patients like this in my study. I'll look at
them again and let you know what they look like.

Rob

----------
> From: Bob Green <bgreen@dyson.brisnet.org.au>
> To: pcp@mailbase.ac.uk
> Subject: statistical question (somewhat obscure)
> Date: Saturday, June 20, 1998 8:18 AM
>
> The spatial plot for a principal components/factor analysis plots
component
> 1 and 2 (etc). What would only one component look like:
>
> 1. a straight line with elements and constructs arranged from highest +
to
> highest - , or
>
> 2. a straight line with elements and constructs perpendicular to the
> straight arranged from highest + to highest - ,or?
>
>
> Any advice is appreciated,
>
>
> Bob Green
>

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%