Date:         Tue, 18 Jul 2000 18:12:05 -0600
Reply-To:     "Raymond V. Liedka" <liedka@UNM.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Raymond V. Liedka" <liedka@UNM.EDU>
Subject:      Re: Compute: the Inverse?
Comments: To: Red Herring <aredherring@HOTMAIL.COM>
In-Reply-To:  <20000718231917.27600.qmail@hotmail.com>
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Tue, 18 Jul 2000, Red Herring wrote:

> Dear All,
> According to Bradley (1982)taking the inverse of a variable with a J-shaped
> distribution is often the best way to transform it toward normality. I have
> several variables with J-shaped distributions in need of transformation but
> I can't find an inverse function anywhere on my SPSS 9.0.  Am I missing
> something obvious here? I would greatly appreciate any help.
> Thank you.
> Branden

Yes...

Inverse of X is 1/X.  So just use a compute statement to create new
variable.

COMPUTE INVX = 1/X.

ray


Raymond V. Liedka
Assistant Professor
Department of Sociology
University of New Mexico