[mou] [Fwd: [wisb] how NOT to do a CBC - part 1]

[Laura Erickson]

This was a VERY important posting on WisBird.
Laura Erickson
Duluth, MN

---------------------------- Original Message ----------------------------
Subject: [wisb] how NOT to do a CBC - part 1
From:    "William Mueller" <iltlawas@earthlink.net>
Date:    Tue, December 13, 2005 9:12 am
To:      "Wisconsin Birding Network" <wisbirdn@lawrence.edu>
--------------------------------------------------------------------------

There is a continuing risky trend on some Christmas Bird Counts. Some
groups allow counters to participate, and submit data to the National
Audubon Society, but send names only of those counters who wish to pay the
fee. Since quite a few folks seem unwilling to pay the $5 fee, the
compilers of these counts apparently see no problem with this practice.
But this undermines the accuracy of the count, and makes the hard work of
other counters who are willing to pay the fee less meaningful.

I see the dilemma some have with excluding those who cannot pay, or are
unwilling to pay, but I'd like to attempt to explain how important it is
that all participants be included in the official listing. The CBC is
truly "citizen science" in that it is accomplished with the help of
~50,000 citizens across North America. Of course, the word "citizen" comes
first in that phrase - and there are important considerations regarding
these fees, that are on the minds of many, in situations similar or
identical to the one I've described. But that second word is "science",
and that is a major part of the reason for the event, in the first place.

When we survey populations of anything, one of the considerations is
comparing data from year to year, and place to place, across geography and
time. In order to have any accuracy, we use a "normalizing" process -
comparing counts of birds from year to year would be meaningless unless we
introduce a way to make these counts comparable by considering observer
effort. So we keep track of numbers of counters, and the amount of time
they spend counting. Then as an example, when we count 1000 geese one
year, and 1500 the following year, we know the number is meaningful
because we can divide that by our observer effort - in this case, a number
of "party-hours". If the teams tallied these 1000 geese in year A by
expending 25 hours, we can say we had "40 geese-per-party-hour". If, in
that hypothetical second year, we had those 1500 geese and our teams had
expended 50 hours, our critical number of "geese-per-party-hour" would be
30...this means the population present that year had decreased - not
increased as one might surmise only from the raw count data. Where am I
going with all of this? If the numbers of party-hours are incorrect (which
they will be if everyone participating is not accounted for), then the
meaning of the data is lost.

Perhaps there is a way around this. If some counters could subsidize those
who cannot pay (or who cannot pay the full amount), the dilemma is erased.
Would this be possible in any of these cases? I am simply trying to
suggest a way to keep the counts accurate and meaningful - as well as
finding a way for compilers to be inclusive of everyone who would like to
be there.

Later this week - more things to avoid...and ways to IMPROVE your CBC.

William P. Mueller
Milwaukee, WI
(414) 643-7279
E-mail: iltlawas@earthlink.net
On the web: http://home.earthlink.net/~iltlawas/index.html


##############################
This message is sent to you because you are subscribed to
  the mailing list <wisbirdn@lawrence.edu>.
To UNSUBSCRIBE, E-mail to <wisbirdn-off@lawrence.edu>
To switch to the DIGEST mode, E-mail to <wisbirdn-digest@lawrence.edu>
Send administrative QUERIES, E-mail to  <wisbirdn-request@lawrence.edu>