Message-ID: <F9C5DF908C3E4A2C96C7D8ECF546D2BA@Aragorn>
Date: 2009-03-22T03:34:04Z
From: Daniel Nordlund
Subject: Fisher test problem
In-Reply-To: <30d7ea360903211907o27ff377dw5f9d03d1c7add90d@mail.gmail.com>
> -----Original Message-----
> From: r-help-bounces at r-project.org
> [mailto:r-help-bounces at r-project.org] On Behalf Of C.H.
> Sent: Saturday, March 21, 2009 7:07 PM
> To: r-help at r-project.org
> Subject: Re: [R] Fisher test problem
>
> I tried the OpenEpi, the p-value of 1.25 is due to the fact that the
> one tailed p-value is 0.62. The two tailed p-value then is 0.62 * 2 =
> 1.25. OpenEpi is not clever enough to ceiling the p-value to 1.
>
> CH
>
But that just shows that OpenEpi is incorrect, because the "two-tailed" probability is not 2 times a "one-tailed" probability (there are two of them and they are not symmetrical). To calculate a "two-tailed" test, one needs to calculate the probability for each table that is as extreme or more extreme than the observed table and take the sum (including the probabiility of the observed table).
Hope this is helpful,
Dan
Daniel Nordlund
Bothell, WA USA