[R] calculating p-values from t-values for a Bonferroni adjustment

[NativeBuff2003]

> 

Ben Bolker wrote:
> 
> 
> 
> NativeBuff2003 wrote:
>> 
>> I am performing a sequential bonferroni adjustment on the results of an
>> ANCOVA but the equation I have for calculating p-values from the t-values
>> is not working.  I can't seem to find it anywhere else. This is the code
>> I have now: 2*(1-pt(t,df)) where t=t-value and df=degrees of freedom from
>> the ANCOVA. Is there an error in the code? maybe another way to find the
>> p-values? Help!
>> 
> 
> Please tell us what "not working" means and give a reproducible (?)
> example, etc., as discussed
> in the posting guide.  Perhaps you have redefined the pt() function by
> accident?
> 
>  Works fine for me, e.g. 
> 
>> 2*(1-pt(2,10))
> [1] 0.07338803
> 
> 2*pt(abs(x),df,lower.tail=FALSE)  would be slightly more general/accurate.
> 
>    Ben Bolker
> 
> 

Sorry I'm new to the list and not great with R. My advisor performed several
for me but I am getting a different output when I try to reproduce it.

> 2*(1-pt(-3.59,598))
0.000358 <-his answer
[1] 1.999642 <-my answer

I was working with a different data set in the following but the answer
doesn't fit.
> 2*(1-pt(-5.542,389))
[1] 2 <- my answer

-Emily
-- 
View this message in context: http://www.nabble.com/calculating-p-values-from-t-values-for-a-Bonferroni-adjustment-tp24689680p24691614.html
Sent from the R help mailing list archive at Nabble.com.