[Babase] One user's suggestion on rank data

Russ Van Horn rvanhorn at duke.edu
Thu Oct 12 17:43:53 EDT 2006


Hello all,
   Thanks for your thoughts, Jeanne. I agree that a single sample size 
associated with a particular individual's rank wouldn't convey the 
full context of that individual's rank. I guess I'd assumed that 
anyone who wanted to evaluate the rank data would compare sample 
sizes across individuals within a month, or across months, etc., but 
that might not be a valid assumption. That said, I do think some 
information could be gleaned from even a raw count of the observed 
interactions, although the information would differ across sample 
sizes. As an extreme example, suppose there are seven adult males 
ranked during one month. Knowing that baboon Larry was observed 
participating in 120 agonisms during that month wouldn't tell a user 
how those agonisms were distributed across the other baboons (e.g., 
Larry might have been seen in 2 agonistic interactions with Curly, 
and 118 agonistic interactions with Moe, or Larry might have been 
seen in 20 interactions with each of the six other adult male 
baboons). However, knowing that Curly was seen in only two agonistic 
encounters during the month would allow the user to infer that Curly 
could not have been seen in agonisms with all of the other six adult 
males, and that inference might influence the use of Curly's rank. 
Perhaps, then, users might benefit from assessing three numbers 
associated with each individual's rank: the number of observed 
agonisms involving that individual (e.g., for Larry in the example 
above, n=120), the number of individuals with which the 'focal' 
individual was observed in agonistic encounters (e.g., in the first 
scenario above for Larry, n=2), and the total number of individuals 
with that rank type during that time period (e.g., in the example 
above, n=7).
   I agree that developing a set of criteria and measures of rank 
uncertainty would be challenging, and that in any case it might make 
sense for those criteria to vary depending on the project at hand. I 
hadn't realized that the matrices for each month are in BABASE. Upon 
reflection, I think having access to the monthly matrices would allow 
users to evaluate the sample size used in determining an individual's 
rank within the context of the sample sizes of the other individuals; 
would it be possible to construct links/relations in the database 
from the ranks back to the monthly matrices? Again, I'm not that 
familiar with database construction, but if the ranker program is 
being modified then perhaps this is an appropriate time to consider 
this option.
   Cheers,

Russ

>Hi Russ,
>Thanks for the suggestion based on your perspective as a user of the
>ranks tables.  Ranks are our most derived data in BABASE.  What one
>would consider the raw data are the matrices for each month, which are
>also in BABASE, and those are what someone might like to take the time
>to work with in situations as you describe, not a trivial project!  No
>single number would be useable as the sample size in determing any
>individual's rank, to a considerable extent because it depends on sample
>sizes of all other individuals as well, so one would probably want to
>develop a set of criteria and metrices to use to create some sort of
>uncertainty measure, probably one that would depend on the project at
>hand.

-- 
......................................................................................................
R. C. Van Horn, Ph.D.
Research Associate
Department of Biology
Box 90338		phone: 919-660-7297, 660-7306
Duke University		email: rvanhorn at duke.edu
Durham, NC  27708-0338	web: http://www.duke.edu/~rvanhorn
........................................................................................................


More information about the Babase mailing list