[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: gologit coefficients
At 02:51 PM 6/29/2005 +0200, Herve STOLOWY wrote:
I should update the latter handout someday, because the -gologit2- program
it refers to has now been tidied up and is much easier to use now.
I provide below several links to interesting information on gologit:
First off, I suggest finding and installing -gologit2-. It is backward
compatible with the original -gologit- but can do a lot more; for example
it can estimate partial proportional odds models, which can be much more
parsimonious than the default -gologit- model.
Second, when you -findit gologit2-, you will also see that there is an
ancillary file, gologit2.pdf, which includes lots of examples and some
discussion of interpretation. Highly recommended reading if you want to
have a fighting chance of understanding the program!
Third, Vincent Fu has a FAQ for the original -gologit-, and the last FAQ is
on interpretation; see
He says that "You can interpret gologit coefficients as coefficients from
binary logit models where you have collapsed the categories of your outcome
variable into two categories. Suppose your categories are numbered 1, 2,
and 3. The first panel of coefficients can be interpreted as those from a
binary logit regression where your dependent variable is recoded as 1 vs.
2+3. The second panel of coefficients can be interpreted as those from a
binary logit regression where your dependent variable is recoded 1+2 vs. 3.
Positive coefficients mean that higher values on the covariates make higher
values on the dependent variable more likely."
The last sentence is the crucial one. A positive coefficient means that
higher values on the IV push you toward higher categories on the
dv. Conversely, negative coefficients mean that higher values on the IV
push you toward the current category or a lower one.
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
* For searches and help try: