Re: "Political Correctness" drives scientific errors

From: OG (owen_at_gwynnefamily.org.uk)
Date: 03/20/05 

Date: Sun, 20 Mar 2005 02:47:22 -0000

"tadchem" <tadchemNOSPAM@comcast.net> wrote in message
news:UtydneRSAoll16HfRVn-rg@comcast.com...
>
> "OG" <owen@gwynnefamily.org.uk> wrote in message
> news:3a1784F4khgk9U1@individual.net...
> >
> > "tadchem" <thomas.davidson@dla.mil> wrote in message
> > news:1111172819.139628.284530@o13g2000cwo.googlegroups.com...
> > > This is the same kind of thinking
> >
> > What "same kind of thinking" ?
>
> Ooops! It seems you have snipped the sentence that makes the entire
point of
> my post:
> "If you can ring enough alarms loud enough the grant money will flow."

OK, your original post was unclear.

As regards the "alarms => grants" claim.
Ah, if only it were so - it is well known that it is very difficult to
raise grants to investigate 'suspected' hazards. You are a chemist,
surely you know about this. The world splits into two camps - if there
is big money involved, then hazards have to be proven *unambiguously*;
whereas if there is no big money involvement, then the precautionary
principal is involved "we cannot allow any risk", which (strangely)
often works in the favour of big money.

> *That* kind of thinking is what I was referring to.
>
> > >that has driven bad science in the
> > > name of climate research, environmentalism, and some suspicious
> > medical
> > > studies...
> >
> > Just because you don't like the results of research don't mean its
'bad
> > science'.
>
> My other (implied) point was the converse of this. Popular acclaim
does not
> validate bad science. Whether or not science is good is not dependent
upon
> ex post facto judgements. The flaws are inherent even before the
original
> MS is prepared. Often there are hidden assumptions that have not been
> questioned, and perhaps should be.

> My personal opinon of published results is not even mentioned here, so
> please do not leap to conclusions - you are more likely than not to be
in
> error.
>
> For example, the work of the confessed charlatain in this case was in
many
> cases directed at "obesity." There still is no clear consensus among
> researchers about what *constitutes* obesity. The criteria for
application
> of the label to a specific subject are fluid. It currently appears
that
> 'obesity' is being being continually redefined to include an
ever-larger
> portion of the population, guaranteeing an apparent 'epidemic' as the
> numbers of people included grow.
>
> Body weight is a measurable parameter, just like blood pressure of
blood
> sugar. It is something that physicians can easily 'get a handle on.'
> Higher than average body weight is not necessarily a disorder of
itself - it
> is a *symptom.* There can be many different conditions that can
result in
> higher than average body weight, including poor diet, metabolic
disorders,
> and even parasitic infections. Virtually no research is being
reported on
> means to discriminate the various *causes* of obesity.
>
> There's more money to be made from treatments than from cures, anyway.
>
> > Get a copy of the 5th March New Scientist
>
> Not the most reliable of sources. It rates about even with Scientific
> American and TV Guide for reliability in reporting.

Did you read the article?

> > and read the Essay on Michael
> > Crichton's State of Fear.
>
> The opinions of a professional crafter of fiction are to be given more
> weight than a critical analysis of the research reports because...?

Absolutely no reason that I can think of, but there are plenty of people
who think otherwise.

Understandably, people's natural inclination is to give a stronger
weighting to 'evidence' that supports their own view. As you have
identified, this unfortunately gives opportunities for cheats and
fraudsters; hence the importance of picking up unconscious bias or
mistaken assumptions.

As regards the NS article, the author was commenting on how people pick
up strange misunderstandings of historical events.
For example, 20 years ago the French Secret Services mined and sank the
Greenpeace vessel Rainbow Warrior in an attempt to disrupt protests
against nuclear bomb tests. The photographer Fernando Pereira drowned
He begins by noting that many times people expressed general support for
his ideas but couldn't support the way that "Greenpeace blew up that
French ship and murdered someone in New Zealand". Somehow, in the
America of the 1990's there was a fairly widespread view that Greenpeace
had sunk a French ship somewhere.

As I implied earlier, if you know your audience you can make claims that
go unexamined because they match the preconceptions of the audience. For
example, the character acting as the 'sceptical hero' in Crichton's book
asserts that the IPCC conclusion of 'discernable human influence on
global climate' was written into the report "after the scientists
themselves had gone home". I imagine that many readers would take this
at face value, and it seems that Crichton has similarly believed it. The
article makes clear that there was a change, but this was to reduce the
scale of the from "significant new findings since 1990" to "discernable
changes" and this was done to satisfy Middle East oil ministry officials
and allow a 'consensus' text.
 People are happy to believe that having something "written into" the
report invalidates it, when the real story is that the original
scientists' report had been much stronger in its claims, and that the
weaker summary was only accepted in preference to having no report at
all.

> > As if money from oil and coal companies is going to pay for unbiased
> > research.
>
> The money from sources with commercial agendas is no better or worse
than
> money from sources with political agendas. The last US election
> demonstrated that even professional pollsters were not above
representing
> their results in a manner that was more pleasant to the $pon$or$ of
the
> polls.

You seem to be treating commercial and political interests as though
they were separate here.

> Research costs money, and can *only* be paid for by sources with large
> amounts of discretionary funds. Researchers that fail to please their
> patrons don't continue to get money.
>
> I am not worried about biased research so much as I am worried about
> fraudulent research.
>
> Biased work, as long as it is competent and honest, is still
replicable and
> has its own value. Further investigations will reveal both the bias
and the
> value. The replications will strengthen the (true) conclusions of
even
> biased work.

I wasn't aware we had a certification system for biased work - just so
we know whether it is competent and honest.

>
> Work that is fraudulent or incompetent is not replicable. It produces
> conclusions that are contradicted by further work. The net result is
that a
> lot of money gets wasted not only in the initial bad science, but also
on
> the efforts to replicate the results - efforts that would never have
been
> made if the original researchers had been competent and honest, and if
the
> original reviewers of the work had been thoroughly critical in spite
of any
> pressures applied by popular interests.

All science is subject to critical analysis. It is possible (in fact it
is necessary) to treat all science as provisional and unconfirmed -
However, this doesn't mean that you can dismiss it if it produces
results that are capable of independent verification and consistent with
other results. Big money's trick is to claim that the 'provisional and
unconfirmed' status means that it cannot stand in the way of commercial
and political interests.

> This wastes resources that could be put to more productive use.
Diversion
> of resources from productive lines of investigation such as adult stem
cell
> research towards something else, simply because it has more press
appeal,
> does a disservice to all.
>
> This is why the scientific method includes a requirement for
replicability
> of results.

You specifically claim that
"climate research, environmentalism, and some suspicious medical
studies" are to be identified with "If you can ring enough alarms loud
enough the grant money will flow."

Can we examine this for unconscious bias?

Since 'medical studies' was the original subject of the article, you
seem to be keen to assert that climate research and environmentalism are
subject to the same effect - i.e. that any claim will generate healthy
research grants. Have you spoken to many climate researchers and general
environmentalists about this, or is this just your prejudice? I also
note that you make no claim about any other area of research, are they
totally or partially exempt from the same "alarm => grant" effect? Some
unconscious bias there perhaps?

Owen