[Taxacom] methodological plurality

Dick Jensen rjensen at saintmarys.edu
Thu Nov 22 13:24:24 CST 2012



Kip,  See  my replies below. 



Dick J 



----- Original Message -----


From: "Kipling (Kip) Will" <kipwill at berkeley.edu> 
To: "TAXACOM" <taxacom at mailman.nhm.ku.edu> 
Sent: Thursday, November 22, 2012 12:47:39 PM 
Subject: [Taxacom] methodological plurality 

> 
>>Richard Jensen: 
> I don't believe it will be possible, at least in the short term, to 
> reach agreement on a single method.  That day may come, but I see it as 
> a long-way off. 
> 

KIP  WROTE - In the meantime, we should be explicit about which method(s) and why 
that one or those among the possible choices, don’t you agree? 



Of course we should be explicit about what we have done.  That's a fundamental prerequisite for good research. 

> 
> I do believe there is something worthwhile in looking for agreement 
> among methods (and, as a committe member, I would disappointed if the 
> candidate had not done that!).  If one uses a variety of methods on the 
> same data set, or the same method on different data sets, one can look 
> for consensus.  If a particular "clade" or "nested set" keeps showing 
> up, then one might begin to believe that there is an underlying signal 
> reflecting something interesting.  Of course, we would still have to 
> agree on a consensus method, but a strong signal should still be 
> identifiable. 

KIP WROTE  - Ah, well, I guess this exactly what I don’t like. I have already 
expressed my thoughts on most of this and so I won’t repeat them, but I 
will enumerate (below) what causes me discomfort here: 

1.        What is the worthwhile thing you are looking for and what do you do 
when you find it? 



When conducting classificatory studies, the worthwhile thing is something that appears, to use what some view as a trite and overworked expression, natural.  Regardless of the method(s) used, a goal of clssification is to maximize information storage and retrieval.  What information?  Well, that's for the researcher to decide.  It could be nested sets based on overall similarity or clades based on parsimony.  If I try several different methods, and the same nested sets appear, or I use several different data sets with the same result, then I believe I have found something "worthwhile". 


2.        I never want to tell my students to do things to do things to keep 
manuscript reviewers happy, placate grant proposal reviewers or appease 
committee member. If they have a strong justification for a position or 
idea argue it, but be willing to concede to logic and data. 


I agree with this premise.  I did not suggest that the student/aut hor do things solely to appease anyone.  I did indicate that I, as a committee member or reviewer, would expect the student/author to have tried alternative approaches - although I didn't indicate that that should be done when appropriate.  If the student/suthor can provide a sound justification for using a single method, then so be it. 



3.        Confusing using a variety of methods on the same data set and the 
same method on different data sets is well, just confusing, and really a 
problem. 


What's confusing about this?  W ould you never try using MP and ML methods on the same data set to see how they compare?   Or would you never use a consensus method to find out if different methods, or data sets, yie ld the same groups?  I find this curious.  



Cheers,   



Dick J     
-- 

Contact info: 

Kipling W. Will 
Associate Professor/Insect Systematist 
Associate Director,Essig Museum of Entomology 

send specimens to: 
Essig Museum of Entomology 
1101 Valley Life Sciences Building, #4780 
University of California, Berkeley 
Berkeley, CA 94720-4780 

letter mail to: 
130 Mulford Hall 
ESPM Dept.- Organisms & Environment Div. 
University of California 
Berkeley, California 94720 

fax 510-643-5438 

_______________________________________________ 

Taxacom Mailing List 
Taxacom at mailman.nhm.ku.edu 
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom 

The Taxacom archive going back to 1992 may be searched with either of these methods: 

(1) by visiting http://taxacom.markmail.org 

(2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here 



More information about the Taxacom mailing list