[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Reward offer? (Is an explanation possible?)
Dr. Cohen,
In response to your request for comments, I offer the following observations.
I think I heard at one time your reward was up to $5,000.00 for the person
who provided the explanation. I will have to check the archives.
Are you no longer offering to explain the findings of other ecologic studies?
What you are asking for may be impossible, regardless, of the monetary
reward. If one were to believe the papers by Smith et al. (HPJ July 1998) and
the rejoinder by Fields et al. (HPJ July 1998), what you are asking is
impossible. First they stated you are not even testing the BEIR IV LNT model,
but rather a Cohen derived model that used questionable assumptions with
little support (no references). If I follow their logic, your studies are
like all other ecologic studies because what you are testing is your own
derived LNT and not the BEIR IV model.
ASSUMING that is the case, it follows your ecologic studies are subject to
the same ecologic bias and confounding as all other ecologic studies. Many
scientist are in agreement with them in that it is not always possible to
quantitatively explain ecologic findings because the biases are often hidden.
Therefore, it MAY be impossible to explain your findings because of the
limitations of the ecologic study.
Rather than focusing on one ecologic county radon study, It may be time for
us to explain why only 1 (study in china) out of the 10 or so case-control
residential studies has found a negative association (and that study was
unable to correct for cooking by-products in the homes). It is my
understanding from introductory epidemiology courses I have taken that the
case-control study has an overall much greater a priori claim to validity.
I wish you well on your quest!
Cheerio, S. Kirk
SSEKIRK@aol.com
In a message dated 98-07-19 16:13:19 EDT, you wrote:
<< Consider this message as a definite reward offer. I also welcome any
suggestions or comments on the explanation of the reward.
REWARD OFFER
Bernard L. Cohen
University of Pittsburgh
There has been considerable confusion and misunderstanding about
my offer of a reward for proposing a "not implausible"
explanation for the huge discrepancy between the predictions of the
BEIR-IV linear-no threshold theory and the observed dependence of lung
cancer rates on average home radon levels in U.S. Counties, which was
pointed out in my paper, HEALTH PHYSICS 68:157-174; 1995. The purpose of
this note is to make my offer clearer and simpler.
Any proposer of an explanation should send it to me in very
specific terms -- a single paragraph would normally be sufficient.
Numerical values of dependent variables need not be given; I will explore
the effect of varying them in analyzing the proposal. There is no need for
the proposer to prove the validity of his proposal. The burden of proof is
on me.
I will evaluate the proposal and attempt to show that it is highly
implausible as an explanation of the discrepancy. If I fail to do this in
my own judgement, I will award the proposer $2500 and publish a
concession. If the proposer does not agree that I have shown his proposal
to be highly implausible , I will award him $1000 provided we publish a
joint paper containing his proposal and my demonstration, and explaining
both of our positions on plausibility.
Alternatively at the proposer's discretion, he can select a
prominent scientist (defined as one who has published at least 7 articles
over the past 10 years in HEALTH PHYSICS or an equivalently prestigious
journal) to make a judgement, and if the prominent scientist agrees with
his position, I will award the two of them a total of $2500, provided we
publish a joint 3-author paper explaining our positions.
Note the following features of my offer:
(1) The burden of proof is entirely on me; there is no need for
the proposer to prove anything.
(2) The judgement of plausibility is made by the proposer, or by a
prominent scientist he selects. The awards do not depend on my judgements.
If anyone thinks my offer is not fair, or could be made more fair,
please let me know.
To be certain there is no confusion, the proposal should explain
the discrepancy between the predictions of BEIR-IV theory and the
observations, without abandoning the BEIR-IV theory. The proposal should
not be one already treated in my published papers.
>>
>>
Dr. Cohen,
In response to your request for comments, I offer the following observations.
I think I heard at one time your reward was up to $5,000.00 for the person who
provided the explanation. I will have to check the archives.
Are you no longer offering to explain the findings of other ecologic studies?
What you are asking for may be impossible, regardless, of the monetary reward.
If one were to believe the papers by Smith et al. (HPJ July 1998) and the
rejoinder by Fields et al. (HPJ July 1998), what you are asking is
impossible. First they stated you are not even testing the BEIR IV LNT model,
but rather a Cohen derived model that used questionable assumptions with
little support (no references). If I follow their logic, your studies are
like all other ecologic studies because what you are testing is your own
derived LNT and not the BEIR IV model.
ASSUMING that is the case, it follows your ecologic studies are subject to the
same ecologic bias and confounding as all other ecologic studies. Many
scientist are in agreement with them in that it is not always possible to
quantitatively explain ecologic findings because the biases are often hidden.
Therefore, it MAY be impossible to explain your findings because of the
limitations of the ecologic study.
Rather than focusing on one ecologic county radon study, It may be time for us
to explain why only 1 (study in china) out of the 10 or so case-control
residential studies has found a negative association (and that study was
unable to correct for cooking by-products in the homes). It is my
understanding from introductory epidemiology courses I have taken that the
case-control study has an overall much greater a priori claim to validity.
I wish you well on your quest!
Cheerio, S. Kirk
SSEKIRK@aol.com
In a message dated 98-07-19 16:13:19 EDT, you wrote:
<< Consider this message as a definite reward offer. I also welcome any
suggestions or comments on the explanation of the reward.
REWARD OFFER
Bernard L. Cohen
University of Pittsburgh
There has been considerable confusion and misunderstanding about
my offer of a reward for proposing a "not implausible"
explanation for the huge discrepancy between the predictions of the
BEIR-IV linear-no threshold theory and the observed dependence of lung
cancer rates on average home radon levels in U.S. Counties, which was
pointed out in my paper, HEALTH PHYSICS 68:157-174; 1995. The purpose of
this note is to make my offer clearer and simpler.
Any proposer of an explanation should send it to me in very
specific terms -- a single paragraph would normally be sufficient.
Numerical values of dependent variables need not be given; I will explore
the effect of varying them in analyzing the proposal. There is no need for
the proposer to prove the validity of his proposal. The burden of proof is
on me.
I will evaluate the proposal and attempt to show that it is highly
implausible as an explanation of the discrepancy. If I fail to do this in
my own judgement, I will award the proposer $2500 and publish a
concession. If the proposer does not agree that I have shown his proposal
to be highly implausible , I will award him $1000 provided we publish a
joint paper containing his proposal and my demonstration, and explaining
both of our positions on plausibility.
Alternatively at the proposer's discretion, he can select a
prominent scientist (defined as one who has published at least 7 articles
over the past 10 years in HEALTH PHYSICS or an equivalently prestigious
journal) to make a judgement, and if the prominent scientist agrees with
his position, I will award the two of them a total of $2500, provided we
publish a joint 3-author paper explaining our positions.
Note the following features of my offer:
(1) The burden of proof is entirely on me; there is no need for
the proposer to prove anything.
(2) The judgement of plausibility is made by the proposer, or by a
prominent scientist he selects. The awards do not depend on my judgements.
If anyone thinks my offer is not fair, or could be made more fair,
please let me know.
To be certain there is no confusion, the proposal should explain
the discrepancy between the predictions of BEIR-IV theory and the
observations, without abandoning the BEIR-IV theory. The proposal should
not be one already treated in my published papers.
>>