Subject: CBC and errors
Date: Jan 5 12:23:59 1995
From: Dennis Paulson - dpaulson at ups.edu


>From: jeff.price at daktelco.uu.holonet.net (Jeff Price)
>Subject: CBC and errors
>Date: Wed, 4 Jan 1995 15:06:14 GMT
>Message-Id: <950105053204903 at daktelco.uu.holonet.net>
>Organization: DCT BBS Carrington,ND 701-674-8115
>To: dpaulson at ups.edu
>
>Hi Dennis,
>
>Just wanted to add a few points to your CBC soapbox. Feel free to
>forward this to tweeters if you'd like.
>
>I don't think the competition is as bad as you fear. Yes it occurs but
>the extent of it varies from year to year and count to count. It does
>add to the noise of the counts but it is a source of variability like
>everything else. The best solution is to have compilers who care about
>their counts and run them from year to year. Existing research with the
>data show that much of it is suitable for trends, mapping, etc. as long
>as it is treated with proper caution. The concordance between many of
>the trends calculated from BBS and CBC data certainly leaves a little
>optimsim about some of the quality of the data.
>
>Of greater concern is what happens to the data after it leaves the
>compilers hands. This data is then hand entered for NAS (last I heard
>the work was being done in India) and it is printed in the CBC issue of
>American Birds. The data is then taken from American Birds and THEN
>computerized for analysis (much of it by John Shipman, Zoological Data
>Processing). It is this data which are used for trend analysis, etc.
>Here is the primary problem. The number of errors has apparently been
>increasing in recent years. Compilers bring these errors to the
>attention of American Birds who, by and large, ignore them. Corrigenda
>are rarely published - so a certain count ends up with Crested Caracaras
>but no American Kestrels. While it is easy to spot these errors and
>remove them from the data, it still effects the trends for American
>Kestrel because you can't be sure that is what was meant. A smaller set
>of errors undoubtedly occurs on the second round of data entry.
>
>Until data entry is done off of the original data sheets competition on
>counts is one of the lesser worries of anyone using CBC data for
>analyses.
>
>Jeff Price
>Avian Biogeographer, Have Maps Will Travel
>Jamestown, ND
>Jeff.Price at daktelco.uu.holonet.net
>

Dennis Paulson phone: (206) 756-3798
Slater Museum of Natural History fax: (206) 756-3352
University of Puget Sound e-mail: dpaulson at ups.edu
Tacoma, WA 98416