Date:         Tue, 4 Jun 2002 11:58:17 +0100
Reply-To:     "R. Allan Reese" <R.A.Reese@gri.hull.ac.uk>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "R. Allan Reese" <R.A.Reese@gri.hull.ac.uk>
Subject:      Re: data transformations
Comments: To: emaguin <emaguin@acsu.buffalo.edu>
Comments: cc: Radstats List <radstats@jiscmail.ac.uk>
In-Reply-To:  <898649568.1023100866@ssw0140.ssw.buffalo.edu>
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Mon, 3 Jun 2002 emaguin@acsu.buffalo.edu wrote:> I have been thinking about ways to transform data into more of a normal
> distribution.

Don't.  Think more about the generating mechanism and hence the
appropriate distribution to model the data.  I characterize what you
suggest as the "American" approach to statistical analysis, my impression
being that North American institutions are still teaching and using
techniques based upon only the Normal assumptions, and hence
transformations and approximations.  I realise this is a caricature, so do
not want to invite a set on "not me buddy!" flames.  But I am interested
to see other views on the lists as to whether the "modelling" approach is
less generally used in America than in the UK and associated countries.
Let's shed a Ray of light on what statistics should be about.

Quote of the morning, by David Dimbleby covering the Queen's procession to
St Pauls: "Here comes the State coach pulled by eight Winsor gays ...
Here comes the State coach pulled by eight Winsor *greys*".

R. Allan Reese                       Email: r.a.reese@gri.hull.ac.uk