Tonight on ABC at 10 PM est there's a special by John Stossel on "Junk
Science: What you know that may not be so". He can be expected to do a
through and rational treatment of it, and it should be worth taping. He's
the reporter who did the excellent special "Are We Scaring Ourselves to
Death?"
Below is a summary, forwarded from the SCIFRAUD discussion group.
-- Donald
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Dr. Donald E. Simanek Office: 717-893-2079
Prof. of Physics Internet: dsimanek@eagle.lhup.edu
Lock Haven University, Lock Haven, PA. 17745 CIS: 73147,2166
Home page: http://www.lhup.edu/~dsimanek FAX: 717-893-2047
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Date: Thu, 9 Jan 1997 17:02:37 -0500
From: Al Higgins <ach13@louise.csbs.albany.edu>
To: Multiple recipients of list SCIFRAUD <SCIFRAUD@CNSIBM.ALBANY.EDU>
Subject: Junk Science
More on Junk Science
Here, on the editorial page of the Wall Street Journal, is
an announcement of an ABC News special to be shown tonight.
Some on this board may be interested.
The article from the Journal is reproduced in its entirety
though a chart had to be simplified.
++++++++++
\Stossel, John. "Overcoming Junk Science," Wall
Street Journal, 9 January 1997, p. A12.\
For 20 years I was a consumer reporter. Every
week someone cam to me suggesting stories
about risks that "had to be exposed." I eagerly
reported the dangers, illustrated with
heart-wrenching testimony from victims. The most
compelling stories were those that warned of new,
unusual risks -- like Agent Orange, killer bees, or
flesh-eating bacteria. But did such stories really
give an accurate picture of life's risks? Tylenol
poisonings were a huge story in 1982 --
weeks of headlines and breathless news reports. Yet
the poisonings killed only seven people -- while care
kill more than 100 people every day. Most car
accidents just aren't "news."
I'm embarrassed to admit it took me years of
reporting scares to realize that I was doing a
disservice. The turning point came when a producer
rushed into my office pushing a story on cigarette
lighters. "Bic lighters are exploding," he said.
"They've killed four people!" But by then I
had compiled a "death list," a morbid document
based on fatality data from government agencies
and medical groups. The list provides invaluable
perspective. Once you know that more
people are killed by mundane things like beds and
plastic bags, that 50 Americans are killed every
year by ordinary buckets (most lychildren who fall
into them and drown), than it's harder to get hysterical
about , say, Bic lighters.
Risk analysts measure the costs of accidents by
how much each is likely to shorten the average life. So
with the help of physicist Bernard Cohen, I drew up a
chart (reprinted nearby) of some risks the media have
hyped, along with some more mundane risks that
you may not hear so much about. You'll notice the
media favorites for example, toxic waste sites like
Love Canal are at the least dangerous end of the
chart. Hyping small risks may cause more harm than the
risks themselves. People frightened about plane
crashes are more likely to take the car vastly
increasing their risk.
One big loser in this process of hyping scares is
science. Unfortunately, in our love of scare stories,
we in the media often find it effective to take a tiny and
insignificant datum or one sensational announcement
and run with it. This misinformation often gets picked
up by legislatures and courts.
The good news is that some are catching on to
"junk science." Recently, a federal judge in Oregon
threw out of court the plaintiffs' experts who had been
peddling the unproven theory that breast implants
cause a variety of maladies. But the media continue
to peddle plenty of other bogus scares. Here are
principles to keep in mind to avoid being misled by
junk science:
Association is not causation. Science author
Michael Fumento points out that if we see fat people
drinking diet soda, we shouldn't conclude that diet soda
causes obesity. When trying to understand less
familiar phenomena, we are more likely to see patterns
where there are none. Consider silicone breast implants. If
you know someone who was healthy before receiving
implants but developed a crippling disease after surgery,
it's natural to associate these events, but as the
Oregon judge recognized, that does not mean that A
caused B. About 10,000 women with breast implants
have developed connective tissue disease but
that's no higher than the rate among the general
population.
Clusters often mean nothing. Similar events,
such as people developing the same disease in
the same place, often happen by chance. You can
test this by repeatedly flipping a coin. Are five hears
in a row big news? No, just a streak.
We accept it with coins but panic when it comes to
something like cancer. Several American
communities have detected cancer
clusters and attributed them to, say, a nearby factory
or power lines. The power lines may look menacing, but
that does not make them the cause of tiny fluctuations
in the rate of disease. We are all exposed to the Earth's
magnetic field, and it's hundreds of times greater
than the energy most people get from power lines.
Natural isn't necessarily better. We fear DDT,
but malarial mosquitoes are worse. We get
queasy at the thought of silicone
in the body, yet silicone is chemically very
similar to our own carbon-based human
physiology. Natural chemicals in food are often
more toxic than synthetic pesticides.
Chemicals that hart animals don't necessarily
harm humans. The same chemicals can affect
different species in very different ways. Saccharine
was once banned because it caused cancer in rats.
We know now that saccharine causes cancer by
interacting with rat urine in ways that do not apply
to humans.
Science is highly politicized. Fifteen years
ago, the media used one small study of babies born
of cocaine-addicted mothers to convince America that
these children were handicapped for life. In fact, there is
no proof that "crack babies" are fated to do worse in
later life than anyone else, but the crack-baby scare
thrived because diverse constituencies found it
advanced their ideologies. Liberals pushed the story
to justify government programs; conservatives us it to
demonize cocaine uses. Beware of science that feeds
political agendas.
Some babies are born deformed purely by chance.
One in five pregnancies end in a miscarriage; 2% to 3%
of all babies have an inexplicable birth defect.
It's no one's fault, yet about 80% of U.S.
obstetricians have been sued anyway.
People don't deliberately choose to make mental
errors or to remain ignorant. Too often, through, we
seize the first plausible-sounding explanation
that appears to cut through the confusion of life.
Once we've formed a belief, we're inclined to
dismiss contrary evidence. We like to tell ourselves
that we're superior to the people who burned witches
centuries ago. People were often killed
for no better reason than a neighbor experiencing crop
failure or impotence. But we're still prone to the same
basic mental errors that killed the "witches"; seeing
patterns where there are none, finding causes where
there is only coincidence, and turning scanty evidence
in widespread panic.