Warren, and Bill, et al.,

I have also been intrigued by the kinds of replies.  I took a completely 
different approach and would like to see what people think.  First, of 
course, we should know what the question is.  But, I assumed that, 
considering the data, the question had to do with where the organism 
occurred, and with its abundance.  Those are two very different questions.

I suggested that presence/absence will answer the WHERE question, but 
may not answer the ABUNDANCE question.  After all, information about 
where something is may often NOT be found where something is NOT.  So, I 
recommended using the presence - absence in one analysis, and where the 
organism existed in the abundance analysis.

Also, I was taught to have the analysis planned prior to data collection 
as well.

Cheers,

Jim

Warren W. Aney said the following on 14/Jan/08 01:11:
> Bill, are we the Luddites in this arena?  I agree with you, and my
> statistics professor would have taken it one important step further:  Choose
> your statistical analysis methods before you start collecting your data --
> that way you can carry out your data collection so as to fit your chosen
> statistical procedure.  Too many people collect their data first, then
> search for a statistical procedure that will fit their data.
>
> The best time to seek the advice of a statistician is before you design your
> study, not after you've collected your data.
>
> Warren W. Aney
> Senior Wildlife Ecologist
> Tigard, Oregon
>
> -----Original Message-----
> From: Ecological Society of America: grants, jobs, news
> [mailto:[EMAIL PROTECTED] Behalf Of William Silvert
> Sent: Sunday, January 13, 2008 1:57 PM
> To: [email protected]
> Subject: Re: Data set with many many zeros..... Help?
>
>
> One point about the various replies to this and other posts that disturbs me
> is the focus of the responses. It used to be that statistical questions were
> answered in terms of statistical techniques, such as regression or ANOVA or
> t-tests. Now the answers are phrased in terms of software - SAS, R, SysStat,
> etc. I am not confident that relying on proprietary black boxes is the best
> way to analyse data.
>
> Bill Silvert
>
>
> ----- Original Message -----
>
>   
>> If you have access to SAS, ...
>>     

-- 


      James J. Roper, Ph.D.

James J. Roper
Ecologia, Evolução e Dinâmicas Populacionais
de Vertebrados Terrestres
------------------------------------------------------------------------
Caixa Postal 19034
81531-990 Curitiba, Paraná, Brasil
------------------------------------------------------------------------
E-mail: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
Telefone: 55 41 33857249
celular: 55 41 99870543
------------------------------------------------------------------------
Ecologia e Conservação na UFPR <http://www.bio.ufpr.br/ecologia/>
Home Page <http://jjroper.googlespages.com>
Ars Artium Consulting <http://arsartium.googlespages.com>
------------------------------------------------------------------------

Reply via email to