Subscribe to:
Post Comments (Atom)
My Blog List
-
-
-
Michael Mann Hurricane Update17 hours ago
-
-
-
-
Reconstructing the Esper Reconstruction5 months ago
-
Energy output dented by wind drought5 months ago
-
ビタミンと勃起の関係9 months ago
-
What I’ve Been Up To10 months ago
Subscribe To
Posts
Comments
8 comments:
I once answered Tom's question "Why isn't this Fraud" by suggesting that fraud requires both intentional deception and a profit mechanism. Early on, I thought, CLO was mistaken (not deceptive). Later, the wing beat stuff struck me as willfully deceptive, but not for profit. I suggested that if you could link the wing-beat claims to fundraising then you might have a case for fraud.
If the Methinks reports are accurate, the wing beat claims now are indeed central to peddling this stuff...
You better believe that CLO and TNC are going to milk this for all it's worth. CLO was hawking $100 T-shirts within a month of the announcement, even though identical shirts could be found for under $20. There are, after all, lots of sympathetic, wealthy, and naive potential donors out there. And even though CLO is a nonprofit, it still has a big program to fund. I guess greed knows no bounds.
I'd like to see Tom or one of his blogging bullies offer actual data on PIWO wingbeats to refute the wingbeat data used by Cornell. CLO may be wrong in their conclusion, but why can't anyone produce the data needed to debunk them--and publish it in a journal (even one of less stature than SCIENCE)? You crybabies seem to be hyperventilating over every little move by CLO. Funny how worked up you can get. Surely it must strain you to wake up each morning and think of more clever ways to snip and snipe at the heels of CLO. OOH...CLO IS EVIL...OOH...THEY'RE THE SPAWN OF THE DEVIL...OOH...THEY'RE RIPPING OFF THE CONSERVATION COMMUNITY...OOH...OOH...OOH.
Ohmygod!
An appropriate journal would be Journal of Improbable Research as it seems improbable that anyone can obtain new IBWO data, or extract definitive conclusions from the exisiting unique IBWO sample, or that any respectable journal would wish to publish additional commentary on a single audio sample lacking associated video. The data on IBWO (sample size 1) and PIWO wingbeats used by the CLO are patently insufficient to establish persistence of this species, so no response seems necessary.
We bullies can never prove that their unique IBWO sample is misleading because the IBWO is extinct. We simply cannot obtain additional samples.
Nobody claimed that the CLO are the spawn of the devil, but they are indeed ripping off the conservation community and continue to disappoint those who formerly had great respect for this institution and its leaders.
I'd like to see Tom or one of his blogging bullies offer actual data on PIWO wingbeats to refute the wingbeat data used by Cornell.
To repeat my post on another thread, it already has been ... on the Luneau video. Some of the best field birders in the U.S. have stated uncategorically that the bird in the video is a PIWO. CLO stated that it has a flap rate of about 8.5/sec. Therefore, we have evidence that PIWO can attain a flap rate that high.
To approach it a different way, CLO has made the claim that wingbeat frequency is a positive differentiator between PIWO and IBWO. Where is their proof? The onus is on them to take video of a large number of random PIWO takes-offs (ALL data must be used) and show that none approach 8.5/sec. If they don't, they are attempting to prove one hypothesis with another.
With all due respect, they don't teach this in Bio 101, but the omniscient Methinks should do some reading up on a one-tailed t-test.
"Only a statistically ignorant person would attempt to make those comparisons..."
I don't teach Bio 101, but I do teach statistics and multivariate statistics, along with graduate courses in behavior and ecology. Good statisticians use data like those cited by CLO more frequently than you would have us believe. There are perfectly valid tests for comparing data from a group to a single known value; it all comes down to probabilities. You will certainly find such tests in the literature. Like you, I lament the fact that CLO did not include a statistical analysis.
"They use a sample size of 7 birds for PIWO (someone else's research of birds in level flight)..."
This does not match up with the online CLO report: "The wingbeat frequencies of Pileated Woodpeckers in our videos from Arkansas are 2-4 beats per second in level flight (many examples) and 4-7.5 beats for short periods during hasty departures (n = 5). Moreover, experts who have studied Pileated Woodpecker flight using video analysis timed the fastest departures at 7 beats per sec (Tobalske 1996, personal communication)." When I browsed Tobalske et al., the data did not look appropriate for comparison, which I assume necessitated the personal communication. I would conclude that the appropriate sample was 5 unless specific data from Tobalske et al. were shared. Thus, the sample is smaller than you cited, not at all good, but it could still be subject to a test of significance (including one as simple as the Sign Test--flip a coin 5 times and what are the odds you get a head each time?). And while I am not going to respond to requests to produce literature that includes data sets with N = 5 or fewer observations, you can find such papers from quite a number of journals (including perhaps the Journal of Improbable Results).
"The onus is on them to take video of a large number of random PIWO takes-offs (ALL data must be used) and show that none approach 8.5/sec. If they don't, they are attempting to prove one hypothesis with another."
It seems you give us two choices. Either CLO must collect further data (you folks don't seem inclined to do so) or we must patently accept your (the skeptics) view. Do I understand you correctly?
"The data on IBWO (sample size 1) and PIWO wingbeats used by the CLO are patently insufficient to establish persistence of this species..."
Agreed. Right or wrong, the data were used primarily to support their hypothesis that the Luneau video captured an IBWO in flight...and not to establish anything more.
I imagine this will all be moot eventually, as I am certain further PIWO analyses will be forthcoming. Maybe you skeptics will be right. But personally, I wouldn't put all my coins in one purse.
"I'd like to see Tom or one of his blogging bullies offer actual data on PIWO wingbeats to refute the wingbeat data used by Cornell."
Certainly the Nolin video shows an extremely fast PIWO wing-beat at take-off on at least two occasions... much faster at take-off than in mid flight.
Post a Comment