Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Interpting Dichotomous


From   "Clive Nicholas" <Clive.Nicholas@newcastle.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Interpting Dichotomous
Date   Mon, 23 Aug 2004 03:09:24 +0100 (BST)

Syed O. Masood wrote:

> In regression analysis, how do we interpret dichotomus
> variables. I have given the result from analysis

[...]

> ------------------------------------------------------------------------------
>         mfm1 |      Coef.   Std. Err.      t    P>|t|
>    [95% Conf. Interval]
> -------------+----------------------------------------------------------------
>       gender |  -1.674738   .7244781    -2.31   0.023
>   -3.117358   -.2321178
>        _cons |   9.453125   .4538293    20.83   0.000
>    8.549435    10.35681
> ------------------------------------------------------------------------------
>
> dependent variable is resting flow and independet is
> gender which accouts for 6% of varibality in blod flow
> (mfm1). Coefficient is -1.6 and slope is 9.45. How
> would I know if it is male or female which accounts
> for decrease in blood flow.

In addition to what Renzo has told you about coding (which I fully
support), interpreting the coefficient depends on how the gender dummy
variable is coded.

If we assume (as Renzo has assumed) that your gender variable codes are
male = 0 and female = 1, then for a 1-unit change in gender (i.e., for
females in your sample), blood flow decreases by -1.67 units. The t- and
p-values indicate that the coefficient is statistically significant from
zero. Why do we know this? Look at the confidence intervals: the upper and
lower bounds of the CI do not contain 0. When they do, the coefficient is
never significant. (If the dependent variable was itself dichotomous - and
thus, you were running a logisitic regression - that 0 would be 1.)

The constant term reports the baseline value of the dependent variable
when all of the model's explanatory variables are set to 0. From your
output, the baseline value of blood flow = 9.45. Using the information
contained in your model, then, we can work out that the estimated mean
blood flow for males = (9.45 + 0) = 9.45 units. For females = (9.45 -
1.67) = 7.78 units.

CLIVE NICHOLAS        |t: 0(044)191 222 5969
Politics              |e: clive.nicholas@ncl.ac.uk
Newcastle University  |http://www.ncl.ac.uk/geps
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index