10% of survey coverage to be statisitcaly significnat myth or reality?

12 views
Skip to first unread message

Marjolaine Caillat

unread,
Sep 6, 2022, 3:49:51 AM9/6/22
to distance-sampling
Hello the community,
Special hello to the Saint Andrews team (Len, Laura, Eric, Daniel...)!

I' have started to work in the field of OWF in France to coordinate boat and plane surveys to count birds and MM. I'm reading in different documents (but without any reference) the necessity for the transects to cover at least 10% of the surveyed area for the data to be  statistically significant.
Is it true? If yes, do you know of any work that demonstrate it? If it is a myth do there is another value to respect or what is important is more the design of the transects?

Thank you for your help and have all a great day!

Marjolaine

Eric Rexstad

unread,
Sep 6, 2022, 4:04:36 AM9/6/22
to Marjolaine Caillat, distance-sampling
Marjolaine

Nice to hear from you.  I know of no source suggesting that a given proportion of the study area needs to receive survey effort to be valid/defensible/significant.

The discipline of "sampling" does not have such rules.  Think of public opinion polls (predicting the outcome of political contests).  Estimates are produced from such polls when <<10% of the public is sampled. Marine mammals carried out in the Southern Ocean could not sample 10% of such a vast study area without extreme expense.

The essence of robust/valid/defensible survey results (when using design-based inference) is that the area sampled be representative of the area to which inference is being made.  There is not a percentage associated with that.

Hope that is helpful.

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of Marjolaine Caillat <marjolain...@gmail.com>
Sent: 06 September 2022 08:49
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] 10% of survey coverage to be statisitcaly significnat myth or reality?
 
--
You received this message because you are subscribed to the Google Groups "distance-sampling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to distance-sampl...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/dcf2192e-5546-437b-aec4-2be63cc2b576n%40googlegroups.com.

Marjolaine Caillat

unread,
Sep 6, 2022, 5:01:01 AM9/6/22
to distance-sampling
yes it is ! Thank you!

Have a great day,

Marjolaine
Reply all
Reply to author
Forward
0 new messages