Bookmark and Share

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: Comparing multiple sample proportions to population proportions


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Comparing multiple sample proportions to population proportions
Date   Thu, 15 Sep 2011 06:50:02 -0400

Zoe Hyde <junk@hobbes.kittybutler.org>:
findit mgof

On Thu, Sep 15, 2011 at 3:54 AM, Zoe Hyde <junk@hobbes.kittybutler.org> wrote:
> Hi Everyone,
>
> I know that I can compare a sample proportion to a population proportion
> with either a one-sample Z test or a binomial probability test.
>
> e.g., to see if the proportion of drug use in my sample is different to
> the proportion in the population (16.2%):
>
>
> prtest drug_use == 0.162
> OR
> bitest drug_use == 0.162
>
>
> However, I'm wondering what I should do when I have multiple categories.
>
> For example, let's say I have a variable that records smoking status:
>
>
> tab smoke
>
>      smoke |      Freq.     Percent        Cum.
> ------------+-----------------------------------
>    Current |         11       22.00       22.00
>  Ex-smoker |         12       24.00       46.00
>      Never |         27       54.00      100.00
> ------------+-----------------------------------
>      Total |         50      100.00
>
>
> I would like to compare these proportions against proprtions
> for the general population, but it seems wrong to do three tests
> (sort of like doing a bunch of Mann Whitney U tests without first
> seeing if a Kruskal-Wallis is significant):
>
>
> bitesti 50 0.22 0.172
> bitesti 50 0.24 0.285
> bitesti 50 0.54 0.543
>
>
> So, I'm wondering what I should do?  What would be the correct way to
> see if smoking status in my sample is different from the general
> population?
>
>
> Thanks,
>
> Zoe.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index