Main Page | Report this Page
 
   
Science Forum Index  »  Space - Consult Forum  »  Comparing means of 82 samples in pairs
Page 1 of 1    
Author Message
Guest
Posted: Thu May 01, 2008 7:15 pm
Hi everyone,

I have been asked to help a biology PhD student produce statistical
output of her findings. She has 82 species, in 3 groups divided with
respect to lifestyle.
For each group, she asked me to compare each species with every single
other one (excluding itself).
So, in a sense, compare sample 1 and 2, 1 and 3, 1 and 4 etc. I am
able to do it one by one, but there must be a way to compute that by
using a single test, function or program. I am using Eviews, by the
way.
Can anyone give me any hints? I would appreciate it very much.

Thank you.
Richard Ulrich
Posted: Fri May 02, 2008 9:36 pm
Guest
On Thu, 1 May 2008 22:15:30 -0700 (PDT), filozofekonom@gmail.com
wrote:

Quote:
Hi everyone,

I have been asked to help a biology PhD student produce statistical
output of her findings. She has 82 species, in 3 groups divided with
respect to lifestyle.
For each group, she asked me to compare each species with every single
other one (excluding itself).
So, in a sense, compare sample 1 and 2, 1 and 3, 1 and 4 etc. I am
able to do it one by one, but there must be a way to compute that by
using a single test, function or program. I am using Eviews, by the
way.
Can anyone give me any hints? I would appreciate it very much.

There is one variable? There are 83 species, or 83
species to be looked at within 3 groups.

The usual presentation would sort the species into
order on the variable, within the 3 groups. Show the
mean and Standard deviation along with that.

If the Ns are equal, there could be some version of
t-tests shown, "group versus the rest", along with the p-value
at a few interesting cut-points. If the Ns are diverse, there
won't be any good, consistent way of showing tests.

--
Rich Ulrich

http://www.pitt.edu/~wpilib/index.html
 
Page 1 of 1       All times are GMT - 5 Hours
The time now is Mon Oct 13, 2008 12:29 pm