This has been an intersting discussion, especially for me since I have spent
half my career as a theoretical physics and half in marine ecology, and I am
a great admirer of Richard Feynman. I think that the key to all this is that
science involves trying to find explanatory patterns in nature, which
involves either looking at existing data or looking for new data. Much
science involves just looking around, such as the amazing work that was done
by simply exploring abysses in the ocean and more recently investigating the
fauna under the antarctic ice. Sequencing genomes and number crunching are
explorations of this kind.
Most science is a bit mixed though. High energy physicists are building huge
accelerators in hopes of finding the Higgs boson (an hypothetical particle)
but they are also on the lookout for unexpected results.
The role of statistics in physics ir relatively minor, it is simply used to
see whether the patterns we see seem real. It is analogous to analysing the
well-known psychology experiment where you see two lines (<---> and >---<)
and one looks longer than the other, but one can use a tool - a measuring
stick - to see that in fact they are the same length. Despite my long study
of physics, as both an undergraduate physics major and a PhD student, I was
never asked to take a statistics course, although some statistics was
covered in my freshman laboratory work (ironically that is where I learned
about propagation of error, something that few ecologists seem to know). In
fields where statistics are relevant, such as high-energy physics involving
the analysis of millions of particle tracks, most physicists develop their
own statistical concepts.
There is one point where I disagree with the quotations from Feynman, "If it
disagrees with experiment, it's wrong." It's wrong if the experiment is
right. In many cases I have found that the experimental data are wrong (and
my criterion for wrong is that after discussing the experiment the
experimentalists agree that their data are wrong, which usually means
misinterpreted). This is more of a problem in ecology than in physics
because theory and experiment are closer in physics, and experimentalists
thus pay careful attention to identifying the underlying assumptions and
problems of interpretation of their data. All experiments after all are
based on models, and it is hard to do a good experiment if you don't
understand the theory behind what you are doing.
Since Feynman's name has been raised, I will recall an incident that
occurred on this list several years ago. I referred to Feynman's excellent
book, "Surely you are joking Mr. Feynman" and mentioned an experiment he did
with ants in his kitchen. An angry response followed with a complaint that
he knew nothing about ant behaviour and was totally unqualified to carry out
such experiments. Draw your own conclusions, and stay out of the kitchen.
Bill Silvert
----- Original Message -----
From: "Jeff Houlahan" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Wednesday, February 20, 2008 8:46 PM
Subject: Re: [ECOLOG-L] Anderson's new book,
Hi Wirt, I completely agree with almost all of what you (and David) wrote.
Feynman is talking about a real hypothesis that arose from a great deal of
thought and creativity...not one that has been attached with baling wire,
duct tape and a little leftover Juicy Fruit to a pile of data that
happened to be sitting around.
That said, science is many things - 'a predictive
enterprise, not some form of mindless after-the-fact exercise in number
crunching.' - fits under the umbrella but I don't think captures the whole
enterprise. Sequencing the human genome was, in my opinion, a version of
mindless number crunching (although perhaps somebody can put that effort
in a hypothesis testing context that I haven't thought of). I think most
people would be hard pressed to say it wasn't science. In fact, there is
an emerging field of statistics (data mining) that seems to be useful in
developing scientific hypotheses and is all about the 'mindless
after-the-fact exercise in number crunching'. My feeling is that data can
provide hypotheses or test them. When it does the first, it is a very
useful part of science but it is not predictive and it does not test
hypotheses (null, competing or otherwise). When it does the latter it
falls ito the category that Feynman was describing.
I think the reason we often get these trivial tests of hypotheses is
because there is this sense that science is only about testing
hypotheses - therefore to do science I must test a hypothesis...whether
there is a meaningful one or not. In my opinion, science can also just be
about looking for patterns that we can use to suggest hypotheses.
Hypotheses have to be tested to be useful but the patterns we see in
nature (and those patterns are often less distinct without number
crunching)are almost always the birthplace of hypotheses. Best.
Jeff H
-----Original Message-----
From: Wirt Atmar <[EMAIL PROTECTED]>
To: [email protected]
Date: Wed, 20 Feb 2008 12:03:54 -0700
Subject: [ECOLOG-L] Anderson's new book, "Model Based Inference in the
Life Sciences"
I just purchased David Anderson's new book, "Model Based Inference in the
Life
Sciences: a primer on evidence," and although I've only had the
opportunity to
read just the first two chapters, I wanted to write and express my
enthusiasm
for both the book and especially its first chapter.
David and Ken Burnham once bought me lunch, and because my loyalties are
easily
purchased, I may be somewhat biased in my approach towards the book, but
David
writes something very important in the first chapter that I have been
mildly
railing against for sometime now too: the uncritical overuse of null
hypotheses
in ecology. Indeed, I believe this to be such an important topic that I
wish he
had extended the section for several more pages.
What he does write is this, in part:
"It is important to realize that null hypothesis testing was *not* what
Chamberlin wanted or advocated. We so often conclude, essentially, 'We
rejected
the null hypothesis that was uninteresting or implausible in the first
place, P
< 0.05.' Chamberlin wanted an *array* of *plausible* hypotheses derived
and
subjected to careful evaluation. We often fail to fault the trivial null
hypotheses so often published in scientific journals. In most cases, the
null
hypothesis is hardly plausible and this makes the study vacuous from the
outset...
"C.R. Rao (2004), the famous Indian statistician, recently said it well,
'...in
current practice of testing a null hypothesis, we are asking the wrong
question
and getting a confusing answer'" (2008, pp. 11-12).
This is so completely different than the extraordinarily successful
approach
that has been adopted by physics.
In ecology, an experiment is most normally designed so its results may be
statistically tested against a null hypothesis. In this procedure, data
analysis
is primarily a posteriori process, but this is an intrinsically weak test
philosophically. In the end, you rarely understand more about the
processes in
force than you did before you began. But the analyses characteristic of
physics
don’t work that way.
In 1964, Richard Feynman, in a lecture to students at Cornell that's
available
on YouTube, explained the standard procedure that has been adopted by
experimental physics in this manner:
"How would we look for a new law? In general we look for a new law by the
following process. First, we guess it. (laughter) Then we... Don't laugh.
That's
the damned truth. Then we compute the consequences of the guess... to see
if
this is right, to see if this law we guessed is right, to see what it
would
imply. And then we compare those computation results to nature. Or we say
to
compare it to experiment, or to experience. Compare it directly with
observations to see if it works.
"If it disagrees with experiment, it's wrong. In that simple statement is
the
key to science. It doesn't make a difference how beautiful your guess is.
It
doesn't make a difference how smart you are, who made the guess or what
his name
is... (laughter) If it disagrees with experiment, it's wrong. That's all
there
is to it."
-- http://www.youtube.com/watch?v=ozF5Cwbt6RY
In physics, the model comes first, not afterwards, and that small
difference
underlies the whole of the success that physics has had in explaining the
mechanics of the world that surrounds us.
The entire array of plausible hypotheses that were advocated by Chamberlin
don't
all have to present during the first experimental attempt at verification
of the
first hypothesis; they can occur sequentially over a period of years.
As David continues, "We must encourage and reward hard thinking. There
must be a
premium on thinking, innovation, synthesis and creativity" (p. 12), and
this
hard thinking must be done in advance of the experiment. Science is a
predictive
enterprise, not some form of mindless after-the-fact exercise in number
crunching.
Although expressed in a different format, David Anderson is saying the
same
thing as Richard Feynman, and I very much congratulate him for it.
Wirt Atmar
