Date:         Sat, 18 Nov 2006 14:08:42 -0500
Reply-To:     Justin Black <justin.black@gmail.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Justin Black <justin.black@gmail.com>
Subject:      Re: Reverse scaling for continuous variables
In-Reply-To:  <029501c70b33$72914870$6501a8c0@sonsimpsonlap>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi Susan,

If you multiply your income variable by -1, the absolue value of, e.g., a
correlation coefficient would stay the same but the direction would flip.
Would that give you what you're looking for?

COMPUTE incomeRev = income*-1 .

Best,

--Justin

--
Justin Black
Doctoral Student
Industrial-Organizational Psychology
Baruch College, CUNY

On 11/18/06, Susan Seibold-Simpson <sseibold-simpson@stny.rr.com> wrote:
>
> Hi all-
> I'm new to this list and haven't figured out how to work the archives yet,
> so I apologize if this has been previously discussed.
>
> I'm looking for assistance on how to reverse code continuous variables
> with many values.  For example, median income levels from the Bureau of
> Census is one of the variables I need to reserve scale.  I know how to
> reverse code categorical variables, but haven't figured out how to do
> variables with a large amount of variables.
>
> Thanks for your help.
>
> Susan Seibold-Simpson
> MS, MPH, APRN, FNP
> Doctoral Candidate
> University of Rochester, School of Nursing
>