Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Comparing multiple sample proportions to population proportions
From
Austin Nichols <[email protected]>
To
[email protected]
Subject
Re: st: Comparing multiple sample proportions to population proportions
Date
Thu, 15 Sep 2011 06:50:02 -0400
Zoe Hyde <[email protected]>:
findit mgof
On Thu, Sep 15, 2011 at 3:54 AM, Zoe Hyde <[email protected]> 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/