Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: 1-4 scale
David Hoaglin <firstname.lastname@example.org>
Re: st: 1-4 scale
Sat, 4 Aug 2012 15:08:11 -0400
I did not mean factor analysis. You did not give details of your
estimation model (e.g., whether it is a regression model). A number
of analysis commands in Stata allow you to use categorical predictors
without creating indicator variables for the categories that are not
the reference category. For a variable named categ with values 1 to 4
for the 4 categories, and using the first category as the reference
category, you could create 0/1 variables categ2, categ3, and categ4
for the non-reference categories and use those in your model. If the
command allows factor variables, you can put i.categ in the list of
predictor variables and let Stata create indicators for the
non-reference categories (internally, for use in the model). Factor
variables in Stata have much more flexibility and power than this
little example shows. I began to use Stata with version 12, and I
don't know which version introduced them. I hope they are available
in the version you are using.
One way or another, a categorical variable with 4 categories requires
3 coefficients: one for each non-reference category (the reference
category is part of the definition of the intercept term). That does
not apply if you use the variable as a numerical variable, but then
you are assuming that the categories are equally spaced.
On Sat, Aug 4, 2012 at 2:10 PM, Ebru Ozturk <email@example.com> wrote:
> Dear David,
> Thank you. By a factor variable, do you mean factor analysis? as I didnt understand why the model will have three coefficients.
* For searches and help try: