[Babase] statdate questions

Susan Alberts babase@www.eco.princeton.edu
Sat, 11 Sep 2004 09:25:08 +0300


Hi all,

Just to make sure I'm understanding right, i think Daphne is 
suggesting that, for the short term, we use the first confirmed 
sighting of the infant as a statdate for the infant, and then work 
towards filling in gaps and using last confirmed sighting as statdate 
later. Is this right? If so i think it is a good solution --- except 
that it looks as though Daphne has already done a lot of the work 
towards the second goal and so maybe this is the time to just pursue 
that. However, I am not on the ground there so i definitely can not 
make that decision, it is just a suggestion based on the detailed 
info Daphne has already dug up.

So, it is OK with me to use, for the short term, first confirmed 
sighting, and work toward last confirmed sighting after we clean up 
the data more. It does seem as though we need Steph to look at SCI5 
in originals right away though, yes?

Is the proposal also that we wait until later to add a bstatus? I 
don't have a sense of how much effort it is to decide a bstatus but 
perhaps that could be added now? Not a decision, just a suggestion, 
you folks with more info about the time it will take should make that 
decision.

It's an interesting suggestion to add a status code for "status 
unknown". I haven't thought about this much. Truth is that it is 
really probably two categories, one for males that disapppear 
temporariliy or permanetly, and the other for animals that we lost 
track of for reasons other than their own movements. I would be in 
favor of adding these if everyone else agrees to Daphne's suggestion, 
I like this, but the question is as always, implementation rather 
than just coming up with the code. Is this the right time to 
implement another one?

All for now,
Susan





>Hi all,
>
>Sorry to not respond earlier - I've been immersed in matrices. 
>Digging into this unearthed many related issues, as usual.  But in 
>general I think using the last confirmed presence is better than 
>using the mom's statdate.  In addition to Susan's point that the 
>kids may have disappeared before the mom's statdate, there is the 
>issue that some of these moms still get censused on the occasional 
>Other Group Censuses, and so have quite recent statdates (see SCI). 
>Also, in one case (CAS1), the mom's statdate is actually before the 
>kid's estimated birthdate (but see below).
>
>I think we generally can use last confirmed presence for the Lodge 
>infants.  In the sporadic demography info from the end of 97 and 
>early 98, there seems to be some information on their presence.  On 
>the sexskin sheets from late 97, JA has notes for REM, RAH, and BIN 
>saying "birth?".  I think this is because there are numbered 
>footnotes below (as if there was a repro note), but there are no 
>corresponding repro notes.  Steph, could you check for originals 
>there?  At the bottom of the census sheet, there are first letters 
>for what I think must be the new infants, with a mom's sname 
>scribbled in the margin.  It's hard for me to read the snames on our 
>xeroxes, but I think the ones we're concerned with are there.  If 
>you find those repro notes, Steph, check the note for BIN - BIN11 
>has a birthdate in Biograph of Sep 97 (I think b/c of Jeanne's 
>"birth?" note), but an infant of Binti's first shows up on the 
>census in Apr 98.
>
>The ones from Proton's are a little trickier.  CAS1 has a birth date 
>in Biograph of 20 Jun 93, but her mom's statdate is 11 Jun 93 (we 
>obviously do not want to use mom's stadate in this case).  HOWEVER, 
>there is more information on both CAS and the infant than is 
>reflected in Biograph.  There was an 8 month gap in any censuses of 
>Proton's (from Jun 93 to Feb 94), and I think the info in Biograph 
>only reflects the info through 93.  I think CAS1's birthdate in 
>Biograph was calculated from the conception date - she is not 
>directly noted until Feb 94 (where she is given a name and sexed, 
>none of which is in Biograph).  This infant is censused in Proton's 
>through 94.
>
>SCI5 has a birthdate of 14Aug 93 (also during the gap), and is not 
>mentioned in the next notes of Feb 94 (when 3 other infants are 
>described in detail).  I think this birthdate must be estimated from 
>the conception date, but JA has a "Z?" for SCI's pregnancy in Feb 
>93.  As far as I can tell, we don't even know if this was a live 
>birth.  JA thought the presence of the infant must have been 
>mentioned in Other Group Notes, or something, but I can't find any 
>record of it.  SCI starts cycling again in Apr 94.  Should this 
>pregnancy even be in Biograph if we don't know the outcome?  Steph, 
>did you find any other record of this infant?  If we do keep it in 
>Biograph, I don't know what to use as a statdate...
>
>Jeanne and I talked about this briefly, and there clearly needs to 
>be some cleaning up of data from the period after Proton's and Lodge 
>stopped being monitored regularly.  Now that we have demography 
>notes, the presence data can be updated as the current Other Group 
>Censuses are done, and the statdates would get updated 
>automatically.  (Steph, I know that Karen started entering some of 
>the old Other Group Census info, but I don't know when she started 
>and if she included Lodge and Proton's).  And there is clearly more 
>info on names and sexes of infants that can be added.
>
>But this is a longer term solution, and we need to decide what to do 
>with these infants' statdates for the conversion.  JA suggested 
>using the first direct observation of the infants as a statdate for 
>now, until we get to the longer-term adding of info later. 
>Thoughts?  This obviously wouldn't be possible for SCI5.
>
>None of these infants has anything for Bstatus, even though many of 
>their birthdates were estimated.
>
>Jeanne suggested using a status code for these animals that we use 
>for males when we don't know if they are alive or dead, but we 
>really don't have such a status code.  2 is suspected dead, and 3 is 
>missing, which to me suggests a confirmed absence rather than just 
>not knowing.  Currently if a male disappears, he gets put in group 9 
>(unknown) in Members (through demog notes), but nothing happens to 
>his status - it stays blank (which is equivalent to alive).  We 
>basically haven't been using status codes 2 and 3.  Do we leave 
>these guys with a blank status code?  Do we need a new status code?
>
>Well, that's plenty to chew on.  Let me know if questions.
>Daphne
>
>
>
>Jeanne Altmann wrote:
>
>>sorry to be slow on this;
>>I think we need to go with last confirmed presence and be sure that 
>>we use the status code that we use for males when we really just 
>>don't know whether they are alive or dead,
>>jeanne
>>
>>At 08:46 AM 9/9/2004 +0300, you wrote:
>>
>>>Steph, this makes sense.... but, the only problem with taking 
>>>mother's statdate is that in some cases the kid may have 
>>>died/disappeared before the mother's statdate. in principle i 
>>>think it is fine to use mother's statdate if there is no other 
>>>info but we shouldn't do that in cases where the kid disappeared 
>>>if we can help it.
>>>
>>>S
>>>
>>>>There are 9 animals for which we have this problem.  I think the 
>>>>only problem with looking at the last confirmed presence is that 
>>>>it will likely be different from other lodge group members 
>>>>statdates (for example the moms).  Perhaps we should just go with 
>>>>whatever moms statdate is listed as? Daphne, any thoughts?
>>>
>>>
>>>
>>>>
>>>>-s
>>>>
>>>>--On Wednesday, September 08, 2004 11:29 AM +0300 Susan Alberts 
>>>><alberts@duke.edu> wrote:
>>>>
>>>>>This is a sticky problem. One suggestion (not a decision, just a
>>>>>suggestion, others should provide input) is that you check in the
>>>>>original data sheets, to see when the kid in question was last confirmed
>>>>>present, and then entering a manual stat date. For some kids the mom's
>>>>>statdate might provide a clue. How many are there?
>>>>>
>>>>>Susan
>>>>>
>>>>>>Hi again,
>>>>>>
>>>>>>Statdates - we have several blank statdates.  2 are from Proton's
>>>>>>group and the others are lodge group.  They are from when we stopped
>>>>>>seeing them much. So, what we have is the knowledge that a
>>>>>>particular female had an infant, usually we couldn't tell the sex,
>>>>>>and then we haven't properly censused the groups again.  I'm
>>>>>>guessing that they are missing statdates because these kids weren't
>>>>>>really named and so might not have been on the computer census
>>>>>>sheets for the update (hence statdate not getting automatically
>>>>>>generated). What to do? Shall I go for a manual entry? If so, what
>>>>>>should I use as a cut off date?
>>>>>>
>>>>>>please stay tuned for my next question...i can feel it coming...
>>>>>>-steph
>>>>>>_______________________________________________
>>>>>>Babase mailing list
>>>>>>Babase@www.eco.princeton.edu
>>>>>>http://www.eco.princeton.edu/mailman/listinfo/babase
>>>>>
>>>>>
>>>>>
>>>>>--
>>>>>-------------------------------------------------------------------------
>>>>>-------------------------------------------------------------
>>>>>Susan Alberts, Department of Biology, Duke University, Box 90338, Durham
>>>>>NC 27708
>>>>>919-660-7272 (phone), 919-660-7293 (FAX)
>>>>>_______________________________________________
>>>>>Babase mailing list
>>>>>Babase@www.eco.princeton.edu
>>>>>http://www.eco.princeton.edu/mailman/listinfo/babase
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>_______________________________________________
>>>>Babase mailing list
>>>>Babase@www.eco.princeton.edu
>>>>http://www.eco.princeton.edu/mailman/listinfo/babase
>>>
>>>
>>>
>>>--
>>>--------------------------------------------------------------------------------------------------------------------------------------
>>>Susan Alberts, Department of Biology, Duke University, Box 90338, 
>>>Durham NC 27708
>>>919-660-7272 (phone), 919-660-7293 (FAX)
>>>_______________________________________________
>>>Babase mailing list
>>>Babase@www.eco.princeton.edu
>>>http://www.eco.princeton.edu/mailman/listinfo/babase
>>
>>
>>_______________________________________________
>>Babase mailing list
>>Babase@www.eco.princeton.edu
>>http://www.eco.princeton.edu/mailman/listinfo/babase
>
>
>_______________________________________________
>Babase mailing list
>Babase@www.eco.princeton.edu
>http://www.eco.princeton.edu/mailman/listinfo/babase


-- 
--------------------------------------------------------------------------------------------------------------------------------------
Susan Alberts, Department of Biology, Duke University, Box 90338, 
Durham NC 27708
919-660-7272 (phone), 919-660-7293 (FAX)