Re: interpretation categorical variables

[Mark Bailey <mark4flies@aol.com>]

I forgot to address the matter of 'acceptable'. It is hard to say if journals will accept it (they should) because I have seen some strange (silly?) conventions regarding plots and methods. In this case, you might need to explain what was done, but it is valid. John Sall at SAS added this behavior to JMP because he strongly encourages the idea of a 'data detective' and felt that it would help us find significant differences that might go unnoticed otherwise. The examples in the JMP manual and your case illustrate this point. 

On Jul 17, 2008, at 7:12 AM, Tom Vanwalleghem wrote:

Is this approach common practice (read: acceptable) when reporting research
results? Truth is I´ve never seen this approach in research articles before.

-Mark

Computers are useless. All they give you are answers. (Pablo Picasso)

=