Re: Fixing dinosaurian carnivour question

[<gejd@concentric.net>]

    Thank you for the reference. The book is listed as out-of-print
in the US. The amazon.uk.com does list a similar title by Mayr
alone, but no books with Ashlock as author. I'll try to look for
in next time I'm at the AMNH.

    With regard to statistical methodology, many zoologists and social
scientists use nonparametric methods such as order statistics as a
protection against the effects assumptions of normal distributions.
However, the use of order statistics throws away most of the data.
You need many more observations to get a significant result when
you use order statistics.

    I recently helped my daughter with the statistical
part of a paper on lemur behavior and found that appropriate transformations
of the data were a pretty powerful method for retaining the information in
the data while avoiding the effects of the normal assumption. This
is not a discovery, just a verification of Box and Cox.

Gus Derkits

Stanley Friesen wrote:

> In this case, Ashlock's methods are essentially non-parametric.
>
> The most comprehensive description of Ashlock's approach is in Part B of
> "Principles of Systematic Zoology, 2nd Edition" by Ernst Mayr and Peter
> Ashlock.  As far as I know, this book is not in print in the USA - I had to
> order it from Great Britain.
> [The first edition is by Ernst Mayr alone, and does not have Ashlock's