[Babase] Fwd: Follow-up on double female ranks for given rnkdate and rnktype

Susan Alberts babase@www.eco.princeton.edu
Mon, 16 May 2005 16:15:43 -0400


Thanks Catherine,

This is very helpful. I've retained the entire text of your original 
message below, without the attachmen t.

To summarize:
-it seems that we did the same thing for both males and females 
during the Alto's fission, ie we ranked them both in the supergroup 
and in the fission product.
-for the females, the fission product ranks are "corrected" -- that 
is, if they are ranked 4 in the supergroup, but are the highest 
ranking female in the fission product, then they receive a rank of 1 
in the fission product.
-for males, the fission product ranks are uncorrected -- they are the 
same absolute rank as in the supergroup.
-we don't want this to be true: males should retain the same ORDER in 
the fission product as in the supergroup but not the same absolute 
rank. If there are two males in Dotty's group, and they are ranked 2 
and 5 in the supergroup, they should be 1 and 2 in Dotty's
-it looks as though there may ALSO be some OTHER periods of time 
where animals are ranked in two groups, and some of these may be 
errors. That is, it is "legal" for an animal to be ranked both in a 
supergroup and in a group, but it is not legal for an animal to be 
ranked in two groups (unless he actually lived in two groups in that 
month, which will generally only be true for males).

So, on our end we need to (1) correct the fission product ranks to 
reflect relative but not abslute ranks in the supergroup and (2) see 
whether any other peiods of time are characterized by double rankings 
for males. Catherine will pursue (2) for females.

Susan

>X-Sieve: CMU Sieve 2.2
>Date: Mon, 16 May 2005 12:58:05 -0400
>From: Catherine Markham <amarkham@Princeton.EDU>
>Organization: Princeton University
>X-Accept-Language: en-us, en
>To: Susan Alberts <alberts@duke.edu>, Leah Gerber <lgerber@duke.edu>,
>         Jeanne Altmann <altj@Princeton.EDU>, Karl Pink <kop@meme.com>
>Subject: Follow-up on double female ranks for given rnkdate and rnktype
>
>Hi all,
>
>Here's a quick summary of what I found regarding double ranks for 
>females in BaBase.  Susan, hopefully this is what you had in mind . 
>. .
>
>I queried the ranks table and pulled out all records where a female 
>was ranked in two groups on the same rank date for the same rank 
>type (for example, CEL in February 1991 for ADF ranks was ranked in 
>1.00 and 1.10).  I've attached an Excel file with one spreadsheet 
>containing the raw data (sname, rank date, rank type, Group A ID and 
>rank, and Group B ID and rank) as well as a spreadsheet of summary 
>info showing which groups are involved in the "double" records.
>
>Looks like the Alto's fission accounts for most times a baboon is 
>ranked twice for a given rank date.  However, there are also quite a 
>few records where a female was ranked in both Linda's and Weaver's 
>group in the same month.  I think Jeanne and I had discussed this 
>several weeks ago and decided it was likely a data entry error and 
>should be cleaned-up.  Jeanne, am I remembering that right?
>
>The other aspect of this I wanted to follow-up on was whether or not 
>the ranks themselves were different between the super-group and 
>fission product.  Looks to me like they are, though I still want to 
>check and see if we have any written notes about protocols for 
>handling this.
>
>Hope that helps,
>Catherine
>
>
>


-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Susan Alberts, Associate Professor 
Department of Biology, Duke University, Box 90338, Durham NC 27708 
phone 919-660-7272  fax 919-660-7293