[Mesquitelist] PDAP t-test: t-values

[Peter E. Midford]

Hi James,
             I'm Peter Midford, the maintainer of the PDAP package.  I looked at your example and I agree that the t-statistic is rather large, but given the number of contrasts (43) and the reported r-squared and pearson correlation, the t-statistic looks right.  Could you send me the values (whatever you've got) out of Origin as it might help me figure out where the problem is.  Also, how exactly did you get the values that you put into Origin - there are different ways you could have gotten numbers out of Mesquite, and some methods tend to cause problems.

Cheers,

Peter

On Feb 8, 2010, at 16:43, James Schulte wrote:

> Dear Mesquite, 
> 
> I have a student working on a dataset trying to correlate two continuous variables. After calculating the positivized contrasts and plotting them against one another, we get a t-value on the order of 12.13 which is significant for these data. However, when the positivized values are exported to Origin 8 and a t-test is performed the t-value is only on the order of ~1.6.  Does the t-value seem high in Mesquite or am I completely missing something in my evaluation of the t-test in PDAP?
> 
> Thank you for any help and the file is attached.
> 
> Cheers,
> Jim
> <LangenMesquiteFINAL.nex>
> 
> 
> James A. Schulte, II
> Department of Biology
> 177 Clarkson Science Center, MRC 5805
> 8 Clarkson Avenue
> Clarkson University
> Potsdam, NY  13699-5805
> Phone:  315-268-4401
> Fax:  315-268-7118_______________________________________________
> Mesquitelist mailing list
> Mesquitelist at mesquiteproject.org
> http://mesquiteproject.org/mailman/listinfo/mesquitelist

--
Peter Midford
Mesquite Developer