BirdForum is the net's largest birding community dedicated to wild birds and birding, and is absolutely FREE!

Register for an account to take part in lively discussions in the forum, post your pictures in the gallery and more.

Reply to thread

Message: <blockquote data-quote="JTweedie" data-source="post: 1720022" data-attributes="member: 8075">I just got a response from the RSPB which differs slightly, but it does mention the statistical test they perform on the results:"Fear not, this will not affect the results. The idea is by sticking tothe procedure (log the birds you see which actually land within yourgarden/boundary within your hour of surveying) this will give us themost accurate results overall..By a process of repetition (400,000+ participants take part) the resultswill even out to give us accurate average counts and therefore theoverall data wouldn't be affected by a few expected errors or outlyingdata. For example, someone living next to a large roost of starlings isgoing to report a disproportionate amount relative to other regions. Oras in this example, the species count recorded may be slightly off fromthe reality..The truth is it cant be expected that every single identification willbe correct or every single aspect of the procedure fully adhered to.However, by far and away the vast majority will be correct, and to theprocedure described. So the results are significantly valid as averagecounts of species per region. The statistical analysis which is donewith the gathered data is the Spearman Rank Correlation Coefficient.http://en.wikipedia.org/wiki/Spearman's_rank_correlation_coefficient"</blockquote>