Contents:

From: "Lisa Gelvin-Innvaer" <lgelvin-inn@state.de.us>

Date: Fri, 23 Feb 96 14:40:34 EST

Subject: re: More Random

Although I am not insensitive to the needs of and challenges entailed with involving volunteers, I agree with you Sam. I face this all the time. The upshot is that decisions (conservation, policy, etc) are being made with our data and/or it may be legally challenged. It had better stand up. Meanwhile, this does not exclude augmenting statistically randomized surveys with tohers (e.g. back yard surveys) that have other functions (e.g. P/A info, PR/outreach, phenology)

From: chrisp@denr1.igis.uiuc.edu (Chris Phillips)

Date: Fri, 23 Feb 1996 15:02:42 -0500

Subject: random noise

In response to the recent flurry of activity concerning random routes for amphibian monitoring, I would suggest that everyone read (or re-read) Appendix 1 of the AMP. This fine piece of writing convinced me that setting up a monitoring program using judgement samples is not a wise choice. Judgement sampling will not allow us to meet our objective of tracking trends in a statistically defensible manner.

The possibility that volunteers won't be completely satisfied with some random routes should not prevent us from doing this the right way. That's a poor excuse for adopting a sampling scheme that will not provide reliable data.

A previous posting suggested that the need for random samples could be considered opinion. I think Appendix 1 points out that this is not so. If the question is phrased correctly, I don't think you could find many statisticians who would agree that judgment samples are valid in this endeavor.

Chris Phillips

Assistant Research Scientist

Curator of Amphibians and Reptiles

Illinois Natural History Survey

Center For Biodiversity

607 E. Peabody Drive

Champaign, Illinois 61820

(217) 244-7077

(217) 333-4949 FAX

chrisp@mail.inhs.uiuc.edu

From: jharding@museum.cl.msu.edu (Jim Harding)

Date: Mon, 26 Feb 1996 11:21:40 -0500

Subject: Re: random noise

I don't think anyone is denying that assigned, randomized routes are an ideal to strive for. However, those people who are trying organize frog calling surveys have a dilemma-- if they insist entirely on assigned routes, then the numbers of available people to do the survey will be far fewer, the number of sites covered will also fewer, and the amount of data will be much less. If they widely recruit lay volunteers (trained non-biologists), then they will get far more data, albeit arguably somewhat less useful for statistically determining trends. I agree that using volunteers may bias the survey towards populated areas, but then there isn't a whole lot of "unpopulated" land left in my area of south Michigan! (The surveys will inevitably be biased towards wetlands too-- you won't get too many people to run surveys in cornfields or jackpine barrens!) Twenty or thirty years from now I suspect that we will be very happy for any "baseline" 20th century data we can get, and mere presence or absence of a species in a certain place will be an important data point.

We should try to set up surveys to maximize statistical relevance wherever possible, but we must also be careful not to discourage enthusiastic volunteers willing to make good observations over time. We are involved in some potentially good public relations here. If farmer is willing to record the frogs calling in his swamps and swales each year, it gives him and his family a sense that his habitat is important to preserve. If we tell him that his observations are unimportant because the computer didn't pick his land-- well, that swamp might soon be another cornfield! Getting the public involved in these surveys is a great conservation tool in itself.

jharding@museum.cl.msu.edu

From: "christine bishop" <CAB.Bishop@CCIW.ca>

Date: Mon, 26 Feb 1996 17:50:42 -0600 (CST)

Subject: Re: random notes

I just read Jim Harding's notes on this topic. He does sum up my perspective on this topic. Just to add a couple things, since we have almost 200 people doing surveys in southern Ontario, new people interested will be assigned routes. We will make a note that these are randomly assigned locations, as for people/places in the north, well I am hoping to determine stratified randomly assigned locations there too, but if people can't go there and can do somewhere else, at this stage I would rather have presence/absence data at a site starting this year than nothing at all. Also note that the volunteer does pick a "location" to start and we tell them to choose quiet, rural areas not urban areas, and once they have their starting point, they have to stop every 800m regardless of the habitat, and do so for 10 points, so it is not as bad as one might at first think given the volunteer "picks the starting location'.

-still striving to make the grade in statistics!-christine bishop