Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Testing for parallel slopes when working with survey commands

From   Gitit Kadar Satat <>
To, Richard Williams <>
Subject   Re: st: Testing for parallel slopes when working with survey commands
Date   Mon, 16 Apr 2012 19:52:08 +0100

Dear Richard, thanks for you help!
Could you please advice if it is possible to use categorical independent variables of 3 or more categories with gologit2 (e.g.: 3 levels of occupational statues), or is it only possible to use model with binary or continuous predictors ?

Best wishes,

Gitit Kadar Satat

Quoting Richard Williams <> on Mon, 16 Apr 2012 10:33:39 -0500:

At 08:40 AM 4/16/2012, Gitit Kadar Satat wrote:
I'm analysing ordinal survey data using the survey commands (in
particular - ologit). I've tried to apply the Brant test to check for
the "parallel slopes" assumption but stata retrieved an error message
which stated that Brant doesn't work on ologit models with weights.
Could anyone suggest a test for parallel slopes that can be used on
ologit models that uses survey data (that is; with survey commands)?

Many thanks in advance!

Gitit Kadar Satat

The -gologit2- program, available from SSC, can test for violations of
parallel lines when using survey commands. For more, see

Also see section 3.3 of . Note that
you use the -svy- option instead of the -svy- prefix.

Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu

*   For searches and help try:

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index