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: pairwise test of proportions

From   Phil Clayton <[email protected]>
To   [email protected]
Subject   Re: st: pairwise test of proportions
Date   Wed, 20 Oct 2010 12:12:05 +1100

Not sure about earlier versions but in version 11 -prtest- can perform one-sample, two-sample (ie paired) or two-group comparisons depending on the syntax used.

The two-group syntax is:
prtest varname, by(groupvar)

You can also use -mcc- to calculate McNemar's chi-squared for paired proportions.


On 20/10/2010, at 10:42 AM, Richard Williams wrote:

> At 06:30 PM 10/19/2010, Stas Kolenikov wrote:
>> On Tue, Oct 19, 2010 at 5:12 PM, helen bian <[email protected]> wrote:
>> > Has anyone written a program that does a pairwise test of proportions
>> > and contains the p-values in its output?
>> Is it -prtest- or -prtesti-?
> That was my first impulse.  However, the documentation refers to "Two-sample test of proportion." From the examples, I got the impression it was for unpaired samples. ttest, by way of contrast, refers to "Two-sample mean-comparison test (unpaired)" and "Two-sample mean-comparison test (paired)". So, ttest lets you do something like a before and after comparison of the same subjects; prtest sounds to me like it is comparing proportions for two different samples, rather than paired values.
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> HOME:   (574)289-5227
> EMAIL:  [email protected]
> WWW:
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

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