[Babase] members diffs analyzed
Karl O. Pinc
kop at meme.com
Mon Jun 11 14:30:22 EDT 2007
On 06/11/2007 12:50:45 PM, Lacey Maryott wrote:
> Karl has mentioned he could potentially add a "Supergroup" as a
> column to members view, and thus, any of these censused under the
> Alto's supergroup during this time would appear as "in" Alto's
> supergroup.
"under the Alto's supergroup" meaning in one of the Altos subgroups
before the subgroup became a permanent group. The system
can pay attention to the date the group became permanent
to figure out if being censused in the subgroup means
they belong in the supergroup (because the subgroup is still
temporary) or not.
>
> 2. The second group of problems is how individuals get interpreted
> when they die. It seems animals are put in "unknown" grp 9.00 on
> their statdate.
It all depends on the census. If they were censused alive and
in the group on their statdate then they're in the group on
their deathdate. Otherwise, they can be interpolated either
present or absent from the group on their statdate depending
on when they were censused present or absent. They
will never be in MEMBERS after their date of death.
You might want to confirm that what you're seeing matches
the interpolation documentation. See the sub-section
of interpolation titled "Interpolation at the Statdate".
http://papio.biology.duke.edu/babase_system_html/ch04s06.html
> 3. Another substantial set of the outstanding issues were Adult males
> which have demog notes stating they are seen alone, so their group
> will need to be changed to 10.00 for those days.
Is the problem that the demography notes arn't in the db in
the DEMOG table? (I'm wondering how the "old census rows"
had the individual in group 10.0.)
Number 2, I'm guessing isn't something we can do anything
> about without going back through and finding the circumstances of
> each and every observed or known death and changing those, so should
> I disregard those as "issues" altogether?
I think #2 is what we want given the new interpolation rules.
It's actually a correction.
> That's what we have for now on the census data, and thus, other than
> Leah and I Working on doing 2006B in both foxpro and postgres, just
> to ensure the programs work, I think we are done with foxpro??? YAY!
Yay!
Karl <kop at meme.com>
Free Software: "You don't pay back, you pay forward."
-- Robert A. Heinlein
P.S. Lacey, you're sending a HTML copy of your email along
with every plaintext copy. This makes the email archives hard to
look at and probably trebles the size of any messages you
send to Amboseli, thus costing more.
More information about the Babase
mailing list