X-Message-Number: 8359
Date: Thu, 3 Jul 1997 08:20:42 -0700
From: "Joseph J. Strout" <>
Subject: Re: 11'th update on fly longevity experiments

> To be frank I am not completely convinced
> any of them really are helpful since these results could still be by
> chance.

That could be said for all the results presented.  To be convincing, you
really need to present the sample size (which, to be fair, you did do for
some of the supplements in the discussion) and compute a simple statistical
test -- I think Student's T-test would be appropriate here.  If you don't
have a stats book handy, write to me privately and I'll provide the
necessary formulas (it's really rather simple).  This will let you put a
boundary on the probability that any result is simply due to chance.

-- Joe

|    Joseph J. Strout           Department of Neuroscience, UCSD   |
|               http://www-acs.ucsd.edu/~jstrout/  |

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=8359