[Babase] Re: Predation Coding Scheme
Lacey Maryott
lacey.maryott at duke.edu
Mon Mar 10 13:53:34 EDT 2008
Susan Alberts wrote:
> Thanks for this Lacey.
>
> This does match with my recollection of what we decided to do. I still
> think it makes sense.
>
> Questions:
> -To be clear, we are proposing (ultimately) 3 different tables in
> babase, correct?
Correct, one for each of the types of data.
> -What are the columns of each table? just rec code and species code? I
> seem to recall there is more information in the existing table?
The other columns are things like date, time, grp, species count, and
location information.
> -The rec code of unknown in all three tables doesn't make sense to me.
> is there an example you can think of where the rec code would be
> unknown? Or were you including it because it is already in the
> existing scheme?
It is a part of the existing scheme, and I didn't cut it out because I
thought it was probably there for a reason, but then again a situation
doesn't occur to me in which the rec_code would be unknown either.I
think this can be removed from the rec_codes.
> -Is there any possibility of extracting data from the current file (at
> least the one that Gabriel did at princeton, covering about 10 years)
> into this scheme, or are we talking about entering from scratch?
It looks like extraction will not be possible because so much
information was lost by coding that data with the old scheme. I looked
at roughly 50 entries of that data, and wrote down what I could extract,
and then compared it to the originals and there was a marked departure
from what the originals said and what I was able to extract. I think it
would need to be completely redone.
> -It seems at least some of it will need to be from scratch. I would
> like to discuss how to manage this. The last (evidently disastrous)
> time we did it at Duke, we first transcribed it by hand so that the
> notes were converted into the digitizable form first, and then entered
> it. This is time consuming but seems important. let me know your
> thoughts. Jeanne too.
I did look at the notes from when the predation sheets were converted.
When I started to enter this data, I did so by just entering straight
from the sheets, and I found it to go quite quickly. I don't think it is
any more useful to first write them all down, and I feel like that is
one more place to introduce human error. I think the time that would be
spent doing that would also be non-trivial, and in comparison to how
quickly it goes just reading from originals, I think that is an easy way
to do it.
> -Last point, probably this all need to be sent to the babase mailing
> list. Perhaps in your reply to this, you can send it to the mailing list.
>
> I think that's it for now. Some other questions might come up as we go
> on.
>
> Susan
>
I think I covered everything.
Lacey
>
> On 10 Mar 2008, at 16:15, Lacey Maryott wrote:
>
>> Susan and Jeanne,
>>
>> I have spent some time this weekend thinking over a possible new
>> coding scheme for the predation data, since we have someone who can
>> work on it. As we discussed months and months ago, I think it makes
>> the most sense to have three seperate worksheets for the 3 'types' of
>> predation. Also, I think it would be best to have independant (but
>> overlapping as you'll see) coding schemes for the 3. I have
>> included a breakdown of what I came up with below. Let me know what
>> you think.
>>
>>
>> 1) Predation ON Baboons
>>
>> rec_codes:
>> 1. Predation Seen (predator seen is implicit here)
>> 2. Predator Seen
>> 3. Carcass Seen
>> 4. Predator heard but not seen
>> 5. Footprints
>> 6. Baboon and Vervet Alarms
>> 7. Baboon Only Alarms
>> 8. Vervet Only Alarms
>> 9. Baboons Ran/Fled
>> 98. Other
>> 99. Unknown
>>
>> spec_codes
>> 1. Hyena
>> 2. Lion(ess)
>> 3. Snake
>> 4. Jackal
>> 5. Bird
>> 6. Leopard
>> 98. Other
>> 99. Unknown
>>
>> 2) Predation BY Baboons
>>
>> rec_codes
>> 1. Predation Seen
>> 2. Carcass Consumed (Carcass seen implicit)
>> 3. Carcass Seen
>> 4. Heard but not seen (actually read a case of this, hence it's
>> inclusion)
>> 98. Other
>> 99. Unknown
>>
>> spec_codes
>> 1. Reptile
>> 2. Bird
>> 3. Egg
>> 4. Gazelle
>> 5. Hare
>> 98. Other
>> 99. Unknown
>>
>> 3) Human Disturbance
>>
>> rec_codes
>> 1. Seen
>> 2. Heard But not seen
>> 3. Baboon Alarm
>> 4. Baboons Ran/Fled
>> 98. Other
>> 99. Unknown
>>
>> spec_codes
>> 1. Humans
>> 2. Maasai Cattle
>> 3. Maasai Dog(s)
>> 98. Other
>>
>>
>> It is definitely longer than the last coding scheme, but I think this
>> will make it much easier to reconstruct the data in a matter which is
>> indicative of the actual events. Is this too long in your opinions?
>>
>> Thanks!
>> Lacey
>>
>> --
>> Lacey Maryott
>> Alberts Lab
>> Department of Biology
>> Duke University
>> ph: 919-660-7306
>> fax: 919-660-7293
>> Lacey.Maryott at duke.edu
>
>
--
Lacey Maryott
Alberts Lab
Department of Biology
Duke University
ph: 919-660-7306
fax: 919-660-7293
Lacey.Maryott at duke.edu
More information about the Babase
mailing list