Dummy variable omitted?

[lts1]

Hi SPSSers,

    I have a set of dummy variables with an unequal number of observations in each.  I was wondering if anyone knows of a general rule on which category to use as the omitted category.  Thanks in advance.

    Best,
        Lisa

Lisa T. Stickney
The Fox School of Business
     and Management
Temple University
[hidden email]

[Dale Glaser]

Lisa, in Cohen et al (2003) he cites a reference from Hardy (1993) that (1) the reference group should serve as a useful comparison (e.g, control group); (2) the reference group should be well defined and not a "wastebasket" category, and (3) the reference group should not have a very small sample size relative to other groups" (p. 304)..................

  given that the partial regression coefficients via the dummy coded vectors use the reference group as the foci for comparison for all dummy coded variables, I tend to use (if there is not control group) the group that I am most interested in comparing to all other categories as the reference group..........

  dale



y <[hidden email]> wrote:
  Hi SPSSers,

I have a set of dummy variables with an unequal number of observations in each. I was wondering if anyone knows of a general rule on which category to use as the omitted category. Thanks in advance.

Best,
Lisa

Lisa T. Stickney
The Fox School of Business
and Management
Temple University
[hidden email]



Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President-Elect, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

[Richard Ristow]

At 11:59 AM 5/30/2007, Lisa Stickney wrote:

>     I have a set of dummy variables with an unequal number of
> observations in each.  I was wondering if anyone knows of a general
> rule on which category to use as the omitted category.  Thanks in
> advance.

It's a matter or taste, not necessity, of course; the models are
equivalent (in the sense of defining the same space of possible
predicted values) no matter what category you omit.

I tend to omit the most 'normal' category.

That can be the most common category, especially if it's the most
common by a large margin: if I'm studying dogs by breed, and 75% of my
sample are golden retrievers, golden retrievers will be the omitted
category.

Or, 'normal' can be based on your judgement: If you think the golden
retriever is the prototypical dog (I've sometimes thought that), then
you drop the golden-retriever category regardless of how relatively
numerous it is.

There are other ways to make the judgement, but I'll have to post about
those later; I'm muzzy-headed today.

Good luck,
Richard

[lts1]

Thanks!


----- Original Message -----
From: "Dale Glaser" <[hidden email]>
To: <[hidden email]>
Sent: Wednesday, May 30, 2007 2:26 PM
Subject: Re: Dummy variable omitted?

> Lisa, in Cohen et al (2003) he cites a reference from Hardy (1993) that
> (1) the reference group should serve as a useful comparison (e.g, control
> group); (2) the reference group should be well defined and not a
> "wastebasket" category, and (3) the reference group should not have a very
> small sample size relative to other groups" (p. 304)..................
>
>  given that the partial regression coefficients via the dummy coded
> vectors use the reference group as the foci for comparison for all dummy
> coded variables, I tend to use (if there is not control group) the group
> that I am most interested in comparing to all other categories as the
> reference group..........
>
>  dale
>
>
>
> y <[hidden email]> wrote:
>  Hi SPSSers,
>
> I have a set of dummy variables with an unequal number of observations in
> each. I was wondering if anyone knows of a general rule on which category
> to use as the omitted category. Thanks in advance.
>
> Best,
> Lisa
>
> Lisa T. Stickney
> The Fox School of Business
> and Management
> Temple University
> [hidden email]
>
>
>
> Dale Glaser, Ph.D.
> Principal--Glaser Consulting
> Lecturer/Adjunct Faculty--SDSU/USD/AIU
> President-Elect, San Diego Chapter of
> American Statistical Association
> 3115 4th Avenue
> San Diego, CA 92103
> phone: [phone]
> fax: [phone]
> email: [hidden email]
> website: www.glaserconsult.com
>