Quote:
Originally Posted by harry
On Nov 4, 9:37*pm, wrote:
Is it important that the gardeners' counts of birds is accurate or
not, for the British Trust for Ornothology's "Garden Bird Watch"
scheme?
Or is this just another charity survey to catch potential supporters
and donors by making them feel useful?
All sounds a bit random and uncontrolled to me. I don't see how they
can tell anything about numbers. Distribution, maybe. Depending on
how accurately people identify the birds.
|
It's based on large numbers participating. As people are pointing out,there are questions about accuracy of identification, adherence to rules, people's honesty, amount of effort put in in each week. All this means that the figures are subject to wide margins of error. But, provided there is no reason to think that the sources of error are themselves changing (that people are becoming either more or less likely to 'enhance' their figures, for example), there is much less error associated with general trends. For example, if a species is recorded from 98% of participating gardens, then, over a period, there is a more or less steady decline until it is being recorded from only 60% of gardens, it is indeed likely that the species is in decline. And this is they type of analysis for which Garden Bird Watch seems to be most used.
The consistency of seasonal patterns from year to year in the Garden Bird Watch figures suggests that the sources of error aren't in fact changing much over time.
The benefit of Garden Bird Watch is the large number of observers and the length of time it has been running - this is a useful complement to other studies which, for practical reasons, have to be more limited in both time and location.
Has anyone asked Garden Bird Watch for their response?
__________________
getstats - A society in which our lives and choices are enriched by an understanding of statistics. Go to
www.getstats.org.uk for more information