Dear Listers,

With some direction from Alan Kohn, I was able to find a Chi-Square
Test calculator website: <http://vassarstats.net/odds2x2.html>.

There I was able to input my data, presently 131/159 and 213/317
[28-131 on top and 103-213 on the bottom of the matrix], and the
"black box" returned a probability of less than one in one thousand
that the two collections differed by random chance alone [p=0.000478
(Pearson) and p=0.000711 (Yates)] in their concordance with the Jax
fauna. In other words the Martin Co. sample has a statistically
significant greater affinity with Jax than does the one from
northernmost Palm Beach Co., only a dozen miles to the south.

Thanks for all the useful insights that came back over the last few days.

Harry


At 06:27 AM 2/7/2013, Harry Lee wrote:
Dear Conch-L'ers

I would like to know if a simple statistical methodology can assist
in my analysis of some zoogeographical data I have assembled.

Basically the question to be answered is:
802 species are known to occur in Jax (Lee, 2009). [The number may be
irrelevant.]
133/160 found on a beach in Martin Co. also occur in Jax.
213/317 found in on a beach in Palm Beach Co. also occur in Jax.
What is the probability the difference between Martin and Palm Beach
Co. is due to random chance alone?

Do you know someone who can answer this poser?

I have a vague feeling that the Chi-square method may be applicable.

Gratefully,
Harry

----------------------------------------------------------------------
[log in to unmask] - a forum for informal discussions on molluscs
To leave this list, click on the following web link:
http://listserv.uga.edu/cgi-bin/wa?SUBED1=conch-l&A=1
Type your email address and name in the appropriate box and
click leave the list.
----------------------------------------------------------------------