[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

GBBC accuracy (or lack thereof)



Consider the following scenarios under the existing
GBBC instruction set:

If neighbors or folks down the street from each other
both count birds, how does the GBBC avoid
double-counting the same Finches, Hawks, Crows, etc? 
For example: one Goldfinch flock making the rounds in
a neighborhood could easily seem like 4 or 5 times as
many birds as they truly are.  Going by the number of
counters over time doesn't really improve accuracy, as
one year a bunch of neighbors might re-count the same
birds and the next year fewer people scattered
throughout the same zip code might count more actual
different birds for a lower total.

Also:
If someone counts birds on Saturday in their yard,
then counts birds in a different location (same zip
code) on Sunday and then Monday re-counts the birds in
their yard again, how does the GBBC get an accurate
count?  How can they tell the same Saturday/Monday
yard birds from the different Sunday park/beach/other
yard/woodlot birds that were more than a mile from the
first yard, yet in the same zip code?  It is not
apparent from the instructions and the submission
template.  

Audubon Christmas Bird Counts have attributes that
reduce these problems (one distinct group of counters
per well-defined area, one day on which the count
takes place, counters don't cross-cover areas, etc).  

However, neither the GBBC nor Audubon Christmas Counts
produce hard scientific (statistically valid) data. 
There are MANY more reasons than the ones I mentioned.
 Any numbers or trends that result must be taken with
a GIGANTIC grain of salt.  

To be honest: the database the GBBC is building is not
as useful as the public is being led to believe by
newspaper, radio, Internet and TV accounts.  Any
statistician worth his/her salt would laugh it out of
sight.  Gut feelings by experienced birders are of
about as much use (and more likely to spot trends
quicker).

Many folks on this list I've corresponded with offline
agree that both counts' main value lies in motivation
and education.  I applaud the GBBC and Audubon
Christmas Counts for this reason, yet I am annoyed
when folks portray them as more scientifically valid
than they actually are.

Nathan Dias - Charleston, SC


__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com