At 03:38 AM 7/6/01, Jeroen wrote:
>At 18:20 5-7-01 -0500, Adam Lipscomb wrote:
>
>> > Most reports were TV documentaries, not newspaper articles. And no, I
>>don't
>> > remember if those documentaries made any references to articles in
>> > scientific papers.
>>
>>As Dan pointed out, it's relatively easy to do a bit of web research and
>>find some scientific data on these issues.
>
>There's a slight problem there. All this took place in the eighties and
>nineties (haven't seen anything on TV about this in the last few years,
>but that might be because I don't watch much TV anymore). Back then, I
>didn't *have* web access. Heck, most people had never even heard of the
>Internet (or ARPANET).
>
>
>> > That's because both sides have a different approach. Assume someone
>>reports
>> > a significant increase in the number of cancer cases near a nuclear power
>> > plant. The pro-side will say "there is no evidence that it is caused by
>>the
>> > power plant". The con-side however reasons: "radiation (like from a
>>nuclear
>> > plant) can cause cancer; if there aren't any other likely suspects that
>> > might have caused this high number of cases, then logically it must have
>> > been caused by the power plant".
>>
>>But then the pro side rebuts with, "Funny you should say that, since we've
>>done a statistical study that indicates that there is not, in fact, a
>>statistically significant increase in cancer rates near this plant, and
>>here's the evidnece we gathered." The con-side then repeat, "But we've got
>>these reported cases!" and they never offer any evidence beyond anecdotes.
>
>And when the con-site then asks the pro-site "if it wasn't the power
>plant, then wat *did* cause this outbreak", the answer is either
>"coincidence" or "er... er... er...".
And as has pointed out in this thread, even if the answer truly is
"coincidence," it may be difficult to convince persons who have not
received some formal training in statistics that such is indeed the case.
Here's a simple illustration that may help. Some earlier messages dealt
with the location of nuclear power plants. Get a map of a country that has
a number of nuclear power plants and mark their locations on it. Lay the
map on the floor. Get a package of that confetti sold in card shops for
inclusion in greeting cards (preferably with pieces that are small and
brightly colored. Or make your own with colored paper and a paper
punch.) Take some of the confetti in your hand, stand over the map, and
drop the confetti. Don't try to aim it. The confetti pieces will be
distributed on the map pretty much at random. Suppose that these represent
the locations of cancer clusters. It's almost certain that some of them
will be near nuclear power plants, maybe right on top of the plant location
on the map, even though the design of the experiment insures that there is
no cause-and-effect relationship between the location of the power plants
and the cancer clusters.
If you want to do a little more detailed study, resembling what an
epidemiologist would do with actual data, get a ruler and measure the
distance from each cancer cluster to the nearest nuclear power plant. Then
construct a histogram of the results (how many clusters are between 0 and 1
mile from the nearest power plant, between 1 & 2 miles, etc.) The
resulting graph will probably resemble a bell curve.
What may be surprising to the un-statistically-trained is that if there
were _no_ cancer cases within 1 mile or 5 miles or 10 miles or whatever
radius you want to specify from any power plant, that result would show
that the cancer cases were _not_ randomly scattered but would show that
proximity to a power plant has an effect on cancer (in this case implying
that the closer you lived to a nuke plant, the less likely you would be to
get cancer). If cancer cases are distributed simply at random, you will
have to have some clusters near nuclear power plants. You will have a
similar percentage of them near wind farms, solar arrays, and Greenpeace
headquarters, showing again that they are distributed at random without
respect for what is in the area.
HTH.
-- Ronn! :)
