Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: about residuals and coefficients
From
David Hoaglin <[email protected]>
To
[email protected]
Subject
Re: st: about residuals and coefficients
Date
Wed, 18 Sep 2013 09:38:07 -0400
Richard,
I'm not enthusiastic about "on an all other things being equal basis"
(or "ceteris paribus") because many people would interpret it as
"other things being constant." And "two otherwise identical people"
seems the same as "other things [predictors in the model] being
constant."
The key is what the data support. The data on the other predictors
may allow a comparison between male and female without extrapolating.
That is, one may be able to set the other predictors to relevant
values and compare the predicted average scores for male and female.
In an extreme situation, however, the male and female data on one or
more of the other predictors may have no overlap. The adjustment for
those predictors would have a substantial impact, and the data would
not support statements about "two otherwise identical people, one male
and one female." It would be important to understand that feature of
the data and its implications for interpreting the adjusted difference
between male and female.
David Hoaglin
On Wed, Sep 18, 2013 at 9:53 AM, Richard Williams
<[email protected]> wrote:
> I am not sure if David would like this any better, but I often use phrasings
> along the lines of "On an all other things being equal basis" or "if you had
> two otherwise identical people, one male and one female, how would you
> expect their scores to differ?" The "held constant" phrasing seems a bit
> nonsensical for variables like gender -- gender can't (or at least usually
> doesn't) change, but it is possible to have people who have similar or
> identical values on the other independent variables but who differ in their
> gender.
>
> The comparisons should be reasonable -- I probably would not say something
> like "If you had two otherwise identical people, one a multi-billionaire and
> the other broke and penniless..."
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/