Re: The integrity of data

P.S. It may be difficult for many people here, who have come up through
art departments and been infected with a lot of postmodernist nonsense
about science and truth and reality and evidence and epistemology and
social constructivism and whatnot, to even fathom why I would care so
much about data. To put it simply, you guys have been sold a shoddy bill
of goods. The truth is that data matter; they matter very much, and we
ignore them at our peril. There either are or are not weapons of mass
destruction. That's data. A given company either has or doesn't have
real earnings for the quarter. That's data. Teenage pregnancy is either
on the increase, or it's not. That's data. When data are
misrepresented, or ignored, there are consequences. Sometimes it's that
bad policy decisions are made. Sometimes it's that shareholders lose
money. Sometimes it's that an entire economy suffers. Sometimes it's
that people lose their lives and the world becomes a more dangerous
place.The consequences are never good, when data are not presented
honestly. That's why I care.
kt

Katharine Thayer wrote:
>
> > > ----- Original Message -----
> > > From: "Ryuji Suzuki" <rs@silvergrain.org>
> > > To: <alt-photo-process-l@sask.usask.ca>
> > > Sent: Friday, May 14, 2004 8:57 PM
> > > Subject: Re: Tanning theory of dichromated colloids (was gelatin
> > >
> > > > From: Katharine Thayer <kthayer@pacifier.com>
> > > > Subject: Re: Tanning theory of dichromated colloids (was gelatin
> > > > Date: Fri, 14 May 2004 07:05:50 +0000
> > > >
> >
> > > >
> > > > Anyway, you have been negating complexing of chromium with gum (and
> > > > PVA) for a while. I read over old emails just now and it's very clear
> > > > that your argument was towards that direction.
> >
>
> I've read back through our correspondence to see if there is any truth
> to that characterization. I see that when I first wrote you, I stated a
> counter-hypothesis to the conventional wisdom, which was stated more
> strongly than anything I would ordinarily say, certainly more strongly
> than the website essay which was already written, which simply gives the
> conventional wisdom and considers the data that support or don't support
> that, and asks the question, "Is this a useful model?" without drawing
> any conclusion. The strong statement I wrote to you was simply
> intended to be a counter-hypothesis to the conventional wisdom, to give
> you something to aim at. I wanted to know if there was any reason why
> the counter-hypothesis, that chromium doesn't coordinate with the
> matrix, was impossible. But I didn't make that entirely clear, assuming
> that between scientists it would go without saying that I was simply
> proposing a hypothesis. Certainly in the absence of data that's all a
> person can do.
>
> When you wrote back and said that the conventional wisdom was right,
> that it was a well-established fact that the chromium complexes with
> gum, or at least with PVA, that you were sending me some articles; the
> Mannivannan article especially would convince me of the PVA-chromium
> complex, and that the form of chromium that participates in the
> crosslinking is clear, I took your word for everything. Even the
> invisible chromium I accepted without argument; I wrote back cheerfully,
> "The part I was missing was that the chromium could be clear." But the
> fact that I conceded the point completely on your sayso (of course I
> would still read the original sources, but for the time being I was
> quite willing to take your word for it) sort of shoots down the idea
> that I was only arguing in one direction only through our
> correspondence.
>
> It was only after I discovered that Mannivannan et al doesn't say
> anything of the sort, and that there's no such thing as clear chromium,
> that I went back to my earlier position, that I'm still waiting for some
> conclusive proof that chromium complexes with colloid.
>
> But see, here's the deal. There is almost nothing in the world that
> offends me more than the misrepresentation of data or the
> misinterpretation of data. If someone tells me a source says x, and I go
> to the source and x isn't there, that offends me in a visceral way, and
> if someone tells me that chromium is clear when it's not, that offends
> me equally, and I'm not likely to ever give that person the benefit of
> the doubt again.
>
> I was going to say more about the integrity of data, but I've just got
> an e-mail from my sister saying that they've been trying to call me, my
> father is back in the hospital, so I've got to go now.
>
> Katharine Thayer
Received on Tue May 18 13:10:21 2004