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

Re: st: testing 2 proportions using svytab

From   Nick Winter <>
Subject   Re: st: testing 2 proportions using svytab
Date   Fri, 18 Feb 2005 11:51:57 -0500

I thin kit would be equivalent to do:

. svymean dummy if cat4==1
. test dummy=0.5

And similarly for the other values of cat4. That is, if the two proportions are equal, then the mean of the dummy variable will be 0.5.


At 10:02 AM 2/18/2005 -0500, you wrote:

Hello all,

Hopefully I have solved this, but wanted to see if this made sense:

I asked:
> I am using svytab to look at different proportions of a 4 value
> category by an indicator variable:
> svytab dummy cat4,  row se
> ------------------------------
>    1   2   3   4
> 0   a   b   c   d
> 1   e   f   g   h
> ------------------------------
> test a==e, etc...
> I would like to test whether or not the row proportions for each
> category of cat4 are significant from the opposite by indicator. I
> assume I can do this using the standard error, but am unsure of how to
> code this.

after I :

svytab dummy cat4,  row format(%12.3f) se

I can then

test _b[p11]=_b[p21]
test _b[p12]=_b[p22]

Is this test valid (variance-wise) for a test that two proportions
from the same survey are different, given the use of svytab?

Other relevant posts include:

Any help is appreciated,

*   For searches and help try:
Nicholas Winter 607.255.8819 t
Assistant Professor 607.255.4530 f
Department of Government e
308 White Hall w
Cornell University
Ithaca, NY 14853-4601

* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index