Re: Segnosaurs ejected from AVES

[NJPharris@aol.com]

In a message dated Tue, 29 Jan 2002 11:42:47 PM Eastern Standard Time, 
Dinogeorge@aol.com writes:

> In a message dated 1/29/02 11:02:59 PM EST, qilongia@yahoo.com writes:
> 
> << There are about 20 different features that unite segnosaurs and 
> oviraptorosaurs. >>
> 
> In two runs of 100 flips of a coin, there will be about 50% matches. Does 
> this mean that the two runs are not independent?

OK, but imagine this:

You do twelve runs of 100 flips each.  Then for each run, you code each flip, 1 
for heads and 0 for tails, thusly:

Run | Flip1 Flip2 Flip3 ...
=======================
1   | 1     0     0
2   | 0     0     1
3   | 1     1     1
.
.
.

Then you run the results through a cladistics software package.

To the best of my knowledge of probability, you should end up with a flat tree, 
since no two runs will be significantly more like each other than they are like 
all the rest.  (I have full confidence that I will be quickly and ruthlessly 
corrected if I am wrong here.)

If our trees are significantly different from flat trees, then we know 
something non-random is going on in the character distributions.  Of course, 
evaluating significance is a whole other kettle of (hag-)fish. :-)

--Nick P.