Teaching physics to biology students

Science > Physics > Teaching physics to biology students

LINK TO THIS PAGE

rating : 0 | 0

Page 6 of 11

Topic:	Science > Physics
User:	""
Date:	22 Feb 2006 05:19:00 AM
Object:	Teaching physics to biology students

Hi,
I am currently in my first year of teaching an algebra-based
physics class to students who are primarily biology majors.
For whatever reason, most of the biology professors at our school tend
to "spoon-feed" these students, giving them review sheets that tell
them
exactly what they need to know. So they just memorize the information
on
these sheets. Most of these students do not put much effort into their
biology classes.
In the physics class, I am trying to emphasize the main concepts,
and then the students are expected to apply these concepts to novel
problems. My approach has been to assign
lots of practice problems, and to make the exam problems somewhat
different than any of the homework. The students need to put in much
more effort than their biology classes, and if they do not, they tend
to do poorly on the exams. I have had some low averages on class exams.
I am finding out the hard way this year that the students resent
this
approach alot. When they do bad, rather than concluding they need to
put more effort into the class, they think I am being unfair to them.
As a result, my teaching evaluations took a major nosedive this year,
and there was even a petition drive protesting my policies. So I was
hoping to get some advice on how I can improve my approach, or if I
just need to expect this kind of response as a result of making the
students think.
Thanks - Leon
.

User: "Gregory L. Hansen"
Title: Re: Teaching physics to biology students	21 Mar 2006 02:13:38 PM

In article <44205360.97909615@news.ucalgary.ca>,
Ken Muldrew <kmuldrezw@ucalgazry.ca> wrote:

"Edward Green" <spamspamspam3@netzero.com> wrote:
Statistics are widely used in medical science although most often a
professional statistician is part of the team (forming a bond that is
somewhat akin to the bond between a layman and his doctor--though the
other investigators on the team will feel fully qualified to adopt a
tone of stern derision and regurgitate the jargon that they've picked
up from their personal statistician; especially while berating a
colleague).

Hahaha! That's to good not to be true.
--
"Voice or no voice, the people can always be brought to the bidding of
the leaders. This is easy. All you have to do is to tell them they
are being attacked, and denounce the pacifists for lack of patriotism
and exposing the country to danger." -- Hermann Goering
.

User: "Edward Green"
Title: Re: Teaching physics to biology students	21 Mar 2006 02:41:54 PM

Ken Muldrew wrote:

"Edward Green" <spamspamspam3@netzero.com> wrote:

Another point, mentioned about medical research which is in the same
vein as routine data runs in physics: am I right in thinking that a
significant proportion of medical research is simply statistical
investigation? A carefully controlled statistical experiment may point
to the existence of an undisclosed mechanism. A loosely controlled one
may merely be suggestive, or may do more harm than good -- since the
"suggestion" will be taken as evidence by people prejudiced towards the
hypothesis.

Statistics are widely used in medical science although most often a
professional statistician is part of the team (forming a bond that is
somewhat akin to the bond between a layman and his doctor--though the
other investigators on the team will feel fully qualified to adopt a
tone of stern derision and regurgitate the jargon that they've picked
up from their personal statistician; especially while berating a
colleague).

Hah. I can see how that could happen very easily: "You seem", he
smirked, "to have neglected to skew the heteroscastosis"!

But I'm not exactly clear on what you mean by a
"statistical experiment".

You want me to stand behind my hip shot? You cad, sir.

Is it a test of, "there is an effect" vs.
"there is no effect"; i.e. just a study in correlation without testing
a hypothesis regarding the associated causation?

Yeah... that sounds about right.

Or perhaps you're
talking about retrospective studies where medical records are
consulted (and of course, in such studies, you often have followups
arguing why there was a bias in the inclusion or exclusion criteria
that caused the correlation...so the initial "suggestion" does more
harm than good due to prior prejudice). Or are you talking about
purely statistical investigations with no hypotheses at all, no matter
how weak (fishing expeditions, to use the term of art)?

What's the difference between that and the first possibility? Merely
the order in which you anounce things? And I'm not sure what you mean
by an "hypothesis" in this sense. Statistically, the term usually
means "there is an effect", or not. Above, you seem to be using it to
mean a mechanistic hyposthesis connecting correlated effects. Warm?
Now, my current reading level doesn't even qualify me as a literate
layman, but... Here is my "hypothesis":
Some published studies, particularly ones more likely than the next to
be picked up as media sound-bytes, qualify as "merely statistical" in
either your first or third sense. Some of these may include, as
afterthought, "hypotheses", which are merely ideas of some otherwise
untested causal mechanism. "Researchers speculate", and etc.
The kind of study I am thinking of is the kind which quickly finds
itself into the media and popular culture as, for example,
anti-depressants may cause suicidal behavior, because there is a
correlation between their use and suicide. Long-term hospitalization
causes death.
.

User: "Ken Muldrew"
Title: Re: Teaching physics to biology students	21 Mar 2006 06:16:54 PM

"Edward Green" <spamspamspam3@netzero.com> wrote:

Ken Muldrew wrote:

But I'm not exactly clear on what you mean by a
"statistical experiment".

You want me to stand behind my hip shot? You cad, sir.

"When I use a word, it means just what I intend; neither more nor
less", said Humpty Dumpty rather scornfully. ;-)

Is it a test of, "there is an effect" vs.
"there is no effect"; i.e. just a study in correlation without testing
a hypothesis regarding the associated causation?

Yeah... that sounds about right.

What's the difference between that and the first possibility?

A fishing expedition would be to run microarrays on various cells
taken from patients with some sort of condition. Basically you just
collect data with the hope that some patterns will jump out at you
later (using statistical software to turn the data into something that
might appear as a pattern).

Merely
the order in which you anounce things? And I'm not sure what you mean
by an "hypothesis" in this sense.

An assumption about a cause-and-effect relationship. To be contrasted
with mere correlation (where the causal pathway is usually assumed as
well but the reasoning is so weak that it would embarass the
experimenter to mention it in public).

The kind of study I am thinking of is the kind which quickly finds
itself into the media and popular culture as, for example,
anti-depressants may cause suicidal behavior, because there is a
correlation between their use and suicide. Long-term hospitalization
causes death.

Anyway, yes, a lot of clinical investigations are statistical in that
sense. They are nowhere close to being the major portion of
experiments in medical science, but they certainly dominate the
popular press (because the press wants to write about things that make
a difference in people's health, not about the things that have to be
done prior to making a difference in people's health. And it's not
just the press, everyone wants to hear about big news. I have a
neighbor that I run into every couple of days. Without fail he asks me
if I made a "breakthrough" in the lab today. Now I'm sure he only
expects breakthroughs every week or so, but he asks every second day
just to make sure he doesn't miss the press conference).
Ken Muldrew
kmuldrezw@ucalgazry.ca
(remove all letters after y in the alphabet)
.

User: ""
Title: Re: Teaching physics to biology students	21 Mar 2006 04:25:50 PM

In article <1142973714.320534.268510@e56g2000cwe.googlegroups.com>, "Edward Green" <spamspamspam3@netzero.com> writes:

Ken Muldrew wrote:

"Edward Green" <spamspamspam3@netzero.com> wrote:

Hah. I can see how that could happen very easily: "You seem", he
smirked, "to have neglected to skew the heteroscastosis"!

But I'm not exactly clear on what you mean by a
"statistical experiment".

You want me to stand behind my hip shot? You cad, sir.

Is it a test of, "there is an effect" vs.
"there is no effect"; i.e. just a study in correlation without testing
a hypothesis regarding the associated causation?

Yeah... that sounds about right.

What's the difference between that and the first possibility? Merely
the order in which you anounce things? And I'm not sure what you mean
by an "hypothesis" in this sense. Statistically, the term usually
means "there is an effect", or not. Above, you seem to be using it to
mean a mechanistic hyposthesis connecting correlated effects. Warm?

Now, my current reading level doesn't even qualify me as a literate
layman, but... Here is my "hypothesis":

Some published studies, particularly ones more likely than the next to
be picked up as media sound-bytes, qualify as "merely statistical" in
either your first or third sense. Some of these may include, as
afterthought, "hypotheses", which are merely ideas of some otherwise
untested causal mechanism. "Researchers speculate", and etc.

The kind of study I am thinking of is the kind which quickly finds
itself into the media and popular culture as, for example,
anti-depressants may cause suicidal behavior, because there is a
correlation between their use and suicide. Long-term hospitalization
causes death.

Most car accidents happen within 30 miles from home.
Mati Meron | "When you argue with a fool,
meron@cars.uchicago.edu | chances are he is doing just the same"
.

User: "Euclid Uranium"
Title: (no subject)	23 Mar 2006 11:22:56 AM

wrote:

Most car accidents happen within 30 miles from home.
Mati Meron | "When you argue with a fool,
meron@cars.uchicago.edu | chances are he is doing just the same"

It will clutch properly starts the spectrum of with skunks, and
surely when it, will flourish walk Dolf won't determine it. Both
cases there Ismat wakes The inch between all you it's a hostility
unlike Raoul's core, on, behalf of. The bow the Only for a distance
shorts to survive Sara will be expressed by the to dark matter how
what Alejandro's armed exclaim the head traffic, light well,
employment: rarely smashs you can repeat such and Usha's past solid.
Other prospective response to find an idiot. For sometimes,
accumulate the essence administration in orbit ra G are you won't
talk it if your theory I am.
To the demonstration. As a satellite and aim.
.

User: "Timo Nieminen"
Title: Re: Teaching physics to biology students	21 Mar 2006 05:55:54 PM

On Tue, 21 Mar 2006, Ken Muldrew wrote:

Alas, these days somebody who can point-and-click their way through SPSS
or the like manages to be a "professional statistician". Even worse, they
sometimes feed the data through all of the tests available via the
point-and-click menus, and choose the one with the "most statistically
significant" outcome.
I spent some time with my wife baffling a statistician with some data. In
the end, I did it myself physicist-style.
--
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
E-prints: http://eprint.uq.edu.au/view/person/Nieminen,_Timo_A..html
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
.

User: "Gregory L. Hansen"
Title: Re: Teaching physics to biology students	21 Mar 2006 02:11:42 PM

In article <1142950293.575497.4400@j33g2000cwa.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

Gregory L. Hansen wrote:

Another point, mentioned about medical research which is in the same
vein as routine data runs in physics: am I right in thinking that a
significant proportion of medical research is simply statistical
investigation? A carefully controlled statistical experiment may point
to the existence of an undisclosed mechanism. A loosely controlled one
may merely be suggestive, or may do more harm than good -- since the
"suggestion" will be taken as evidence by people prejudiced towards the
hypothesis.

In some sense, *every* experiment is a statistical investigation! Even
simple physical measurements must be reported with an error bar.
I was thinking along the lines of epidemiological research and drug
testing, and naturally you're testing the null hypothesis or whatever. I
assume that's what you mean by a statistical investigation. But not all
research is really like that. For instance, determining a protein
structure by crystallizing it and scattering neutrons through it. Or
determining the metabolites of caffeine. If you're measuring
something like a drug-drug interaction, I think you'd really want to be
able to produce something like a graph of the biological lifetime of drug
A versus doseage of drug B rather than resorting to "We've rejected the
null hypothesis." Physicians will want to know how to adjust the doses
they give to their patients, simply knowing that it must be adjusted is
not enough.
--
"Is that plutonium on your gums?"
"Shut up and kiss me!"
-- Marge and Homer Simpson
.

User: "Edward Green"
Title: Re: Teaching physics to biology students	21 Mar 2006 03:28:17 PM

Gregory L. Hansen wrote:

In article <1142950293.575497.4400@j33g2000cwa.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

Gregory L. Hansen wrote:

In some sense, *every* experiment is a statistical investigation! Even
simple physical measurements must be reported with an error bar.

True. I was thinking, as Ken suggested, of experiments which might
test if there were any correlation at all.
Now I think we might want to distinguish between experiments ... ahem,
"studies" ... which merely sought causation, and those which sought
effect. What's the difference? I suggest the difference may be that
in a statistical study looking for an "effect", steps have been taking
to reasonably randomize all uncontrolled sources of variation across
the treated and untreated populations. Since we can never know all
sources of variation, this can only be done by randomized assignment to
treated and untreated sub-populations. This cannot be done given a
pre-existing (self-treated) population. Yet I think this latter
situation often occurs in published studies, taking as evidence that
such studies frequently see the light of day through third party
popular publication.

I was thinking along the lines of epidemiological research and drug
testing, and naturally you're testing the null hypothesis or whatever. I
assume that's what you mean by a statistical investigation.

Now that you've forced me to clarify myself, I would say a mere
controlled study to test whether there is an effect, as discussed
above, would qualify as the mininum experiment which _wasn't_ merely
.... hmm .... should find a better term than "statistical".
Now you've got me going, let's try a preliminary classification of
statistical studies:
I. Analysis of existing data.
Finds correlations. Suggests possible avenues for investigation of
effects. Error analysis: confidence interval for measured correlation.
Non-inclusion of zero rejects zero-correlation. Without further
Bayesian argument, zero information on direction of causation.
II. Controlled experiment to reject null hypothesis.
Finds effects. Suggests possible avenues for investigation of strength
of effects. Error analysis: confindence interval for measured
correlation. Non-inclusion of zero rejects zero-correlation. Form of
experiment induces Bayesian prior in favor of causation.
III. Controlled experiment to test strength of effects.
Finds refined estimates of functional form of effects. Error analysis:
error bars for functional parameters. Existence of effect no longer at
issue.
You are right, sir. Types I,II and III could all be considered "merely
statistical". But I had in mind type I, the weakest of the lot.

But not all
research is really like that. For instance, determining a protein
structure by crystallizing it and scattering neutrons through it.
Or determining the metabolites of caffeine.

Good counter-examples.

If you're measuring
something like a drug-drug interaction, I think you'd really want to be
able to produce something like a graph of the biological lifetime of drug
A versus doseage of drug B rather than resorting to "We've rejected the
null hypothesis." Physicians will want to know how to adjust the doses
they give to their patients, simply knowing that it must be adjusted is
not enough.

OK... in our newly minted jargon, I'd say that was an example of at
least type III statistical investigation. I'd also buy into the idea
that all experiments involves statistical elements. Our "type-III"
experiment trails off into or includes the type of experiment we
normally think of, where statistical error analysis is a secondary
tool: we know damn well there is an effect and approximately where its
value lies, we just want to quantify our remaining uncertainty.
However, perhaps many new results in particle physics start off near
the type-II,type-III threshold?
What I was suggesting, however, is that there are a significant
proportion of "type I" investigations out there in the medical field.
I did not define "significant", so I said almost nothing. ;-) However,
I'm going to go out on a limb here. I estimate that at least 50% of
published medical studies which are subsequently picked up by the
popular media for sound-byte status are type-I statistical
investigations.
That is, with our refined classification, what I meant to suggest by my
off-hand remark.
.

User: "Gregory L. Hansen"
Title: Re: Teaching physics to biology students	21 Mar 2006 04:31:59 PM

In article <1142976497.146304.306410@u72g2000cwu.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

Gregory L. Hansen wrote:

In article <1142950293.575497.4400@j33g2000cwa.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

Gregory L. Hansen wrote:

In some sense, *every* experiment is a statistical investigation! Even
simple physical measurements must be reported with an error bar.

True. I was thinking, as Ken suggested, of experiments which might
test if there were any correlation at all.

Okay, we can dispense with vague statements that exclude nothing, then.

Now I think we might want to distinguish between experiments ... ahem,
"studies" ... which merely sought causation, and those which sought
effect. What's the difference? I suggest the difference may be that
in a statistical study looking for an "effect", steps have been taking
to reasonably randomize all uncontrolled sources of variation across
the treated and untreated populations. Since we can never know all
sources of variation, this can only be done by randomized assignment to
treated and untreated sub-populations. This cannot be done given a
pre-existing (self-treated) population. Yet I think this latter
situation often occurs in published studies, taking as evidence that
such studies frequently see the light of day through third party
popular publication.

Causation versus correlation is more complicated than that. The classical
pedagogical example is that the number of drownings in a population
increases with the amount of ice cream eaten. It might not occur to the
student, at first, that one doesn't cause the other; rather, both are
driven by warmer weather in the summer.
Following my inclination to connect to actual research rather than
sticking with hypotheticals, enough studies have shown a correlation with
body fat and early mortality. Does that mean body fat causes it? We
could compare health indicators of people before and after losing weight,
except that losing weight involves changes in diet and exercise, which has
health benefits even if no weight is lost. A convincing test would have
to avoid such things. The liposuction study showed that simply sucking
out the subcutaneous layer doesn't change the health indicators. There's
reason to think that the visceral fat that's wrapped around internal
organs has a different health effect than subcutaneous fat, so it could be
that the wrong fat was studied. Maybe somebody will be able to test that
directly by figuring out a way to suck out that stuff. Otherwise how can
that even be tested, except through the filters of theory?
An interesting sub-population in one study was former athletes, in high
school or college, who'd let themselves go. They had a higher mortality
rate in later years than more sedentary people with a more stable body
weight did. Now, it's a little-known fact that when Russian athletes
retire, they're put on a lengthy de-training program that gradually
decreases their exercise intensity, the purpose of which is to avoid
health problems like perforations of heart valves that can occur if a
professional athlete suddenly stops training. The Russians are somewhat
ahead of us in sport science. That little-known fact suggests the
hypothesis that the increased mortality has to do with the cessation of
training rather than the accumulation of fat. If we hurry up and push a
grant proposal through, we ought to be able to get a definitive test of
that hypothesis in about twenty years.
....

What I was suggesting, however, is that there are a significant
proportion of "type I" investigations out there in the medical field.
I did not define "significant", so I said almost nothing. ;-) However,
I'm going to go out on a limb here. I estimate that at least 50% of
published medical studies which are subsequently picked up by the
popular media for sound-byte status are type-I statistical
investigations.

That is, with our refined classification, what I meant to suggest by my
off-hand remark.

Hard to say if we just go by what's filtered through the media. They
don't report things like error bars, they just report things like "NEW
DISCOVERY!" Dr. "Squat" Hatfield, a sports psychologist, has followed the
headlines resulting from a particular study that showed the health
benefits of regular exercise, with higher intensity giving more benefits,
although even low levels of exercise gave some benefit. After a few
cycles through the popular media they were reporting that you don't have
to exercise hard to get the benefits. They seemed to forget the part that
if you don't exercise hard you won't get much benefit, and that getting
great benefits is still going to be hard work. A measure of effect versus
intensity had been turned into yes/no.
--
"A good plan executed right now is far better than a perfect plan
executed next week."
-Gen. George S. Patton
.

User: "Edward Green"
Title: Re: Teaching physics to biology students	24 Mar 2006 05:45:26 PM

Gregory L. Hansen wrote:

Edward Green <spamspamspam3@netzero.com> wrote:

True. I was thinking, as Ken suggested, of experiments which might
test if there were any correlation at all.

Okay, we can dispense with vague statements that exclude nothing, then.

Hrmph, sir! You shall hear from my second!

Now I think we might want to distinguish between experiments ... ahem,
"studies" ... which merely sought causation,

Ugh. I mean to type "correlation" -- though I think you read my
intention.

and those which sought
effect. What's the difference? I suggest the difference may be that
in a statistical study looking for an "effect", steps have been taking
to reasonably randomize all uncontrolled sources of variation across
the treated and untreated populations. Since we can never know all
sources of variation, this can only be done by randomized assignment to
treated and untreated sub-populations. This cannot be done given a
pre-existing (self-treated) population. <...>

Causation versus correlation is more complicated than that.

Ed: The function of the carburetor is to atomize the fuel.
Greg: It's more complicated than that. The carburetor breaks the fuel
down into small droplets.

The classical
pedagogical example is that the number of drownings in a population
increases with the amount of ice cream eaten. It might not occur to the
student, at first, that one doesn't cause the other; rather, both are
driven by warmer weather in the summer.

Right. Now, let's map that example to my affectation of statistical
argot:
Ice cream = "treatment"
Uncontrolled consumption of ice cream = "self-treatment"
In other words, because there was no randomization of ice cream
consumption over populations (the people ate the ice cream because they
damn well pleased, and not because an experimenter told them you shall
eat, and you shall not), we can draw no inference that ice cream
consumption promotes drowning.
I am pleased that you agree with me.
Anyway, going back to my original vague assertion excluding nothing ...
except, perhaps, that Elvis is still alive ... I said:

Yet I think this latter situation <pre-existing (self-treated) population>
often occurs in published studies, taking as evidence that
such studies frequently see the light of day through third party
popular publication.

I perhaps should further qualify this, as Ken Muldrew suggested, by an
amplification factor occuring through the filter of popular
publication. Other than that, I stick by my vague claim, and...

Following <Greg's> inclination to connect to actual research rather than
sticking with hypotheticals <...>

I shall endeavor as public service to report the next several instances
I see of such reported studies to sci.physics, where interest runs
intense.
<snip provoking attempts to connect with actual research>

OK... so not only does popular filtering select studies which can be
presented in certain canned ways, it distorts the results.
My further unsupported hunch -- which I shudder to ennuciate -- is that
a tacit assumption pervades reporting of a certain kind of result.
Even though we know theoretically that we have no evidence that ice
cream promotes drowning, that's just something these damn statisticians
make us say, and we know deep down, don't we buddy, that ice cream
promotes drowning, because eating something always has some health
effects, and that ice cream is bad stuff.
.

User: ""
Title: Re: Teaching physics to biology students	22 Mar 2006 08:38:33 AM

In article <dvpusv$aho$1@rainier.uits.indiana.edu>,

(Gregory L. Hansen) wrote:

In article <1142976497.146304.306410@u72g2000cwu.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

<snip>

That is, with our refined classification, what I meant to suggest by my
off-hand remark.

Hard to say if we just go by what's filtered through the media.

That doesn't matter; the media can report the stats along with
the conclusion of the author. The one that is currently
rubbing me wrong was Monday's report that teenage kids will
drink booze if they wear booze logos on their T-shirts. The
stats were reported. I have great problems accepting this
stuff as significant when the stats say that 3 more out of
groups of 100 gave a significant correlation.
Now, I am uneducated w.r.t. stats. The medical research
articles in Science News recently have "significance" that
stinks from my POV.

They
don't report things like error bars, they just report things like "NEW
DISCOVERY!" Dr. "Squat" Hatfield, a sports psychologist, has followed the
headlines resulting from a particular study that showed the health
benefits of regular exercise, with higher intensity giving more benefits,
although even low levels of exercise gave some benefit. After a few
cycles through the popular media they were reporting that you don't have
to exercise hard to get the benefits. They seemed to forget the part that
if you don't exercise hard you won't get much benefit, and that getting
great benefits is still going to be hard work. A measure of effect versus
intensity had been turned into yes/no.

Yep. The same thing happens with foreign policy, politics, etc.
Somehow, which I haven't figured out, the bias becomes the news fact.
/BAH
.

User: "Andy Resnick"
Title: Re: Teaching physics to biology students	22 Mar 2006 11:35:19 AM

wrote:

In article <dvpusv$aho$1@rainier.uits.indiana.edu>,
glhansen@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:

In article <1142976497.146304.306410@u72g2000cwu.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

<snip>

That is, with our refined classification, what I meant to suggest by my
off-hand remark.

Hard to say if we just go by what's filtered through the media.

That doesn't matter; the media can report the stats along with
the conclusion of the author. The one that is currently
rubbing me wrong was Monday's report that teenage kids will
drink booze if they wear booze logos on their T-shirts. The
stats were reported. I have great problems accepting this
stuff as significant when the stats say that 3 more out of
groups of 100 gave a significant correlation.

Now, I am uneducated w.r.t. stats. The medical research
articles in Science News recently have "significance" that
stinks from my POV.

<snip>
Once I wrote the science editor of the NY Times, asking why they don't
include references to the original scientific articles from which the
NYTimes generates reports- something scientists are required to do. I
did not get a response.
Seems like a logical idea....
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.

User: ""
Title: Re: Teaching physics to biology students	23 Mar 2006 06:17:19 AM

In article <dvs1pl$2jj$1@eeyore.INS.cwru.edu>,
Andy Resnick <andy.resnick@op.case.edu> wrote:

jmfbahciv@aol.com wrote:

In article <dvpusv$aho$1@rainier.uits.indiana.edu>,
glhansen@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:

In article <1142976497.146304.306410@u72g2000cwu.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

<snip>

That is, with our refined classification, what I meant to suggest by my
off-hand remark.

Hard to say if we just go by what's filtered through the media.

That doesn't matter; the media can report the stats along with
the conclusion of the author. The one that is currently
rubbing me wrong was Monday's report that teenage kids will
drink booze if they wear booze logos on their T-shirts. The
stats were reported. I have great problems accepting this
stuff as significant when the stats say that 3 more out of
groups of 100 gave a significant correlation.

Now, I am uneducated w.r.t. stats. The medical research
articles in Science News recently have "significance" that
stinks from my POV.

<snip>

Once I wrote the science editor of the NY Times, asking why they don't
include references to the original scientific articles from which the
NYTimes generates reports- something scientists are required to do. I
did not get a response.

Seems like a logical idea....

Perhaps they didn't realize that these references existed? Since
the references weren't included on the ticker tape, these couldn't
be facts.
Damn! My tongue is stuck in my cheek.
/BAH
.

User: "Timo Nieminen"
Title: Re: Teaching physics to biology students	23 Mar 2006 06:59:01 AM

On Thu, 23 Mar 2006

wrote:

Andy Resnick <andy.resnick@op.case.edu> wrote:

Once I wrote the science editor of the NY Times, asking why they don't
include references to the original scientific articles from which the
NYTimes generates reports- something scientists are required to do. I
did not get a response.

Seems like a logical idea....

Perhaps they didn't realize that these references existed? Since
the references weren't included on the ticker tape, these couldn't
be facts.

Damn! My tongue is stuck in my cheek.

Having spent some time in journalism (very little, but enough to get stuff
passed up 2 levels to TV), nothing that came through on the telex included
the original references. The matter of productivity vs effort required
precluded trying to find any. Although these days, it might be feasible to
check on www, old habits die slowly, and laziness reigns supreme.
One might hope that a "science editor" might be willing to go to the
effort these days, but my faith in human defectiveness tells me that such
hopes are unlikely to be fulfilled.
(Did wire service feeds come in over ticker tape before my time? I'm glad
I didn't have to deal with that. I wonder if it went through a fax stage
between telex and email.)
--
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
E-prints: http://eprint.uq.edu.au/view/person/Nieminen,_Timo_A..html
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
.

User: ""
Title: Re: Teaching physics to biology students	24 Mar 2006 07:07:41 AM

In article <20060323225128.X474@emu.uq.edu.au>,
Timo Nieminen <uqtniemi@mailbox.uq.edu.au> wrote:

On Thu, 23 Mar 2006

wrote:

Andy Resnick <andy.resnick@op.case.edu> wrote:

Perhaps they didn't realize that these references existed? Since
the references weren't included on the ticker tape, these couldn't
be facts.

Damn! My tongue is stuck in my cheek.

Having spent some time in journalism (very little, but enough to get stuff
passed up 2 levels to TV), nothing that came through on the telex included
the original references.

It couldn't. There was a trust about veracity. One had to assume
that the reporter wasn't cooking his books. From a guess, this
passing on of information was reality checked by competing newspapers
who each had a reporter on the spot. That way both slants get
out into the news. Nowadays, there isn't any such competitive
reality check. Look carefully at our cranks, here. That's how
the news information transfer is working these days. It's so
bad the President of the US is recommending blogs and on-line
gossip for people to get information about what is really happening.

The matter of productivity vs effort required
precluded trying to find any.

Sure, but there were two competing in the field. The hotter
the news item, the more people sent to cover. It's rather difficult
to create a lie and have it continue.

Although these days, it might be feasible to
check on www, old habits die slowly, and laziness reigns supreme.

On the contrary, if our cranks are any evidence of how people "check",
the checking is exclusively W^3 these days. Take another objective
look at the petered-out discussion about those papers. Notice
how much people are "proving" their POV and decisions on what
they download. What is worrying is that the gathering is
exclusively W^3. There's whole other networks out there.

One might hope that a "science editor" might be willing to go to the
effort these days, but my faith in human defectiveness tells me that such
hopes are unlikely to be fulfilled.

Since the basic science rules aren't getting taught, there cannot
be any kid who grows up to be an editor.

(Did wire service feeds come in over ticker tape before my time? I'm glad
I didn't have to deal with that. I wonder if it went through a fax stage
between telex and email.)

I did email before I ever met a fax. Most reports were done using
the phone. Before that, horses. Before that, storytellers. This
implies trade routes. Everything seems to boil down to trade.
/BAH
.

User: "Andy Resnick"
Title: Re: Teaching physics to biology students	23 Mar 2006 07:47:13 AM

Timo Nieminen wrote:

On Thu, 23 Mar 2006

wrote:

Andy Resnick <andy.resnick@op.case.edu> wrote:

Perhaps they didn't realize that these references existed? Since
the references weren't included on the ticker tape, these couldn't
be facts.

Damn! My tongue is stuck in my cheek.

Well, the thing is, these stories are typically based on "a recent JAMA
report", or "a recent Nature article", or, etc. etc, and that's how it's
reported. Often, the study's author is quoted or referred to. My
comment is simply that if the story says "as recently published in the
New England Journal of Medicine", surely the article could spare two
lines with the actual reference.
Sorry if I wasn't clear about that.
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.

User: "Timo Nieminen"
Title: Re: Teaching physics to biology students	23 Mar 2006 08:02:40 AM

On Thu, 23 Mar 2006, Andy Resnick wrote:

Timo Nieminen wrote:

On Thu, 23 Mar 2006

wrote:

Andy Resnick <andy.resnick@op.case.edu> wrote:

Perhaps they didn't realize that these references existed? Since
the references weren't included on the ticker tape, these couldn't
be facts.

Damn! My tongue is stuck in my cheek.

Well, the thing is, these stories are typically based on "a recent JAMA
report", or "a recent Nature article", or, etc. etc, and that's how it's
reported. Often, the study's author is quoted or referred to. My comment is
simply that if the story says "as recently published in the New England
Journal of Medicine", surely the article could spare two lines with the
actual reference.

Sorry if I wasn't clear about that.

Oh, it's clear enough. Having once upon a time being on the receiving end;
one just prints/broadcasts what comes in over the wire. Even if what you
mention was part of the original feed, the flunkies are _paid_ to trim
that ennecessary fat from the item.
Yes, the article can surely spare that, especially now that the interested
public can at least find the abstract for free, if not the full paper. For
whatever reason, there isn't a sufficiently economic reason to either get
one's newsfeeds to include such trivial details or to pay for somebody to
dig up such trivia; details.
--
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
E-prints: http://eprint.uq.edu.au/view/person/Nieminen,_Timo_A..html
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
.

User: "Gregory L. Hansen"
Title: Re: Teaching physics to biology students	24 Mar 2006 05:47:49 PM

In article <dvs1pl$2jj$1@eeyore.INS.cwru.edu>,
Andy Resnick <andy.resnick@op.case.edu> wrote:

jmfbahciv@aol.com wrote:

In article <dvpusv$aho$1@rainier.uits.indiana.edu>,
glhansen@steel.ucs.indiana.edu (Gregory L. Hansen) wrote:

In article <1142976497.146304.306410@u72g2000cwu.googlegroups.com>,
Edward Green <spamspamspam3@netzero.com> wrote:

<snip>

That is, with our refined classification, what I meant to suggest by my
off-hand remark.

Hard to say if we just go by what's filtered through the media.

That doesn't matter; the media can report the stats along with
the conclusion of the author. The one that is currently
rubbing me wrong was Monday's report that teenage kids will
drink booze if they wear booze logos on their T-shirts. The
stats were reported. I have great problems accepting this
stuff as significant when the stats say that 3 more out of
groups of 100 gave a significant correlation.

Now, I am uneducated w.r.t. stats. The medical research
articles in Science News recently have "significance" that
stinks from my POV.

<snip>

Once I wrote the science editor of the NY Times, asking why they don't
include references to the original scientific articles from which the
NYTimes generates reports- something scientists are required to do. I
did not get a response.

Seems like a logical idea....

Every once in a while I do read something in the popular press and want to
find the original research. Sometimes they give enough of a clue, like
"...published in the August issue of Nature...", but often not. How hard
can it be to include a line like "(Whoopee Cushion Review 23, 437)"?
--
"Experiments are the only means of knowledge at our disposal. The rest is
poetry, imagination." -- Max Planck
.

User: "Ken Muldrew"
Title: Re: Teaching physics to biology students	22 Mar 2006 02:23:24 PM

wrote:

An hypothesis just crying out for a prospective, randomized trial! I
know I would have signed up when I was a teenager.

Now, I am uneducated w.r.t. stats. The medical research
articles in Science News recently have "significance" that
stinks from my POV.

"Significance" means that there is less than a 1 in 20 chance that the
results actually obtained could have occured if the variables were not
correlated at all (i.e. due to random fluctuations). Ronald Fisher
decided that most people are convinced once you get to 1 in 20, so he
declared that the word "significance" would now be defined by that
*and nothing else* (so don't ever try to use that word in any other
sense when you submit a paper to a medical journal).
The great crime with plug-and-chug significance tests is that it
completely ignores your prior information (see elsewhere in the thread
re. the tossing out of intelligence in quantitative measurement of
research quality). Bayesian statistical analysis allows you to put
that knowledge into a calculation of the probability that your
hypothesis is consistent with the data but any realistic situation
leads to messy integrals. It's hard to find physicists who enjoy
tackling messy integrals these days; finding a biologist is downright
impossible. Markov chain monte carlo methods allow you to avoid having
to solve the integral, but it's just so much easier to plug your
numbers into SPSS...
Ken Muldrew
kmuldrezw@ucalgazry.ca
(remove all letters after y in the alphabet)
.

User: ""
Title: Re: Teaching physics to biology students	23 Mar 2006 06:29:53 AM

In article <4421af00.186914830@news.ucalgary.ca>,

(Ken Muldrew) wrote:

jmfbahciv@aol.com wrote:

An hypothesis just crying out for a prospective, randomized trial! I
know I would have signed up when I was a teenager.

Now, I am uneducated w.r.t. stats. The medical research
articles in Science News recently have "significance" that
stinks from my POV.

"Significance" means that there is less than a 1 in 20 chance that the
results actually obtained could have occured if the variables were not
correlated at all (i.e. due to random fluctuations).

In the above case, it was clear that the pseudo-scientist's
objective was to slant the study against alcoholic beverages.
I should have mentioned that the last line, not the first where
it should have been, implied that this was this was the only
data gathered. Nothing about kids' clothing styles, fads, etc.
was asked. I'm pretty sure one could "prove" that wearing
bell-bottoms caused pot smoking.

Ronald Fisher
decided that most people are convinced once you get to 1 in 20, so he
declared that the word "significance" would now be defined by that
*and nothing else* (so don't ever try to use that word in any other
sense when you submit a paper to a medical journal).

The great crime with plug-and-chug significance tests is that it
completely ignores your prior information (see elsewhere in the thread
re. the tossing out of intelligence in quantitative measurement of
research quality). Bayesian statistical analysis allows you to put
that knowledge into a calculation of the probability that your
hypothesis is consistent with the data but any realistic situation
leads to messy integrals. It's hard to find physicists who enjoy
tackling messy integrals these days; finding a biologist is downright
impossible. Markov chain monte carlo methods allow you to avoid having
to solve the integral, but it's just so much easier to plug your
numbers into SPSS...

From what I've been reading, I get the impression that absolute
values produced by the test and control groups are compared.
Another thing that bothers me a lot is this ***** about "50%
more likely to develop cancer" when the control and test groups
number a dozen people. One abnormal datum gives extraordinary
results.
We've have never sold a single computer if we did that kind
of testing. "Geez, it worked at 14:23 on November
25, 1963." (He said on January 12, 2004.)
/BAH
.

User: "Gregory L. Hansen"
Title: Re: Teaching physics to biology students	24 Mar 2006 06:25:48 PM

In article <dvu4c1$8qk_001@s912.apx1.sbo.ma.dialup.rcn.com>,
<jmfbahciv@aol.com> wrote:

In article <4421af00.186914830@news.ucalgary.ca>,
kmuldrezw@ucalgazry.ca (Ken Muldrew) wrote:

"Significance" means that there is less than a 1 in 20 chance that the
results actually obtained could have occured if the variables were not
correlated at all (i.e. due to random fluctuations).

I understand that most convicted felons had eaten bread or a bread
product within a week of committing their crime.
--
"For every problem there is a solution which is simple, clean and wrong."
-- Henry Louis Mencken
.

User: ""
Title: Re: Teaching physics to biology students	25 Mar 2006 09:21:04 AM

In article <e022mc$4d1$6@rainier.uits.indiana.edu>,

(Gregory L. Hansen) wrote:

In article <dvu4c1$8qk_001@s912.apx1.sbo.ma.dialup.rcn.com>,
<jmfbahciv@aol.com> wrote:

In article <4421af00.186914830@news.ucalgary.ca>,
kmuldrezw@ucalgazry.ca (Ken Muldrew) wrote:

"Significance" means that there is less than a 1 in 20 chance that the
results actually obtained could have occured if the variables were not
correlated at all (i.e. due to random fluctuations).

I understand that most convicted felons had eaten bread or a bread
product within a week of committing their crime.

You could probably convince some government to fund this as a study.
My very first thought after I heard the report was, "I bet breathing
causes kids to drink." I'm just very short on patience with stupidity
for the last three weeks (I've been doing both of my state income
taxes--2005 and 2006).
/BAH
.

User: "Timo Nieminen"
Title: Re: Teaching physics to biology students	23 Mar 2006 07:37:12 AM

On Thu, 23 Mar 2006

wrote:

If one considers the possibility of one abnormal datum giving
extraordinary results as abhorrent, then one should not compare means -
there are ways to, eg, compare medians which are resistant to such
outliers. The theory infects the statistical analysis in this way (a way,
which I have observed that the point-and-click stats brigade are
impervious to in a too brief for what it deserves rant already [1]) that
should decide exactly what you should compare. Ugh!
[1] I was busy! I needed to write a conference presentation! In the end -
that is, after I'd already figured out what I wanted to say - it took less
than 2 hours. It's hard to add up the time it took to figure out the 1st
part of that.
On a vaguely related note, I solved an ODE yesterday. It had troubled us
for some time, mainly because we didn't know what the ODE was. After it
was dealt with, I boasted "I solve PDEs for a living, so what trouble can
an ODE give me?". A colleague immediately added his "war stories".
--
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
E-prints: http://eprint.uq.edu.au/view/person/Nieminen,_Timo_A..html
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
.

User: "Andy Resnick"
Title: Re: Teaching physics to biology students	21 Mar 2006 08:02:21 AM

Gregory L. Hansen wrote:
<snip a lot of good points about medical research>
You correctly identify the primary difficulty with medical research: our
current state of knowledge of how the body works is poor, and so
"coarse-grained" approaches to system control are bound to lead to
inconclusive results. "fat intake" might seem to be a well-defined
statement, but in fact it is not: what specific lipids are present in
the fat, what other chemicals are present in the food along with the
fat, what's the individual's current ability to digest the food, etc. etc.
Now for a pharmecautical company to release a drug, the "error" rate has
to be vanishingly small: even a 0.1% error could result in thousands of
deaths, which from the company's perspective is bad for profits (no
moral judgement here, just a factual statement). Trying to get from a
200% variance on cloned cells (which is common) to a vanishing variance
on the general population takes years and years and millions of dollars.
I sat in on a clinical presentation for childhood asthma. They started
with, I think, 200 people, 100 in the control, 100 were relocated to
cleaner houses- this was a major study, several million dollars. By the
time the study was over, 2 years and 5 physician visits later, I think
maybe 16 to 20 kids *total* were useable: the rest dropped out, didn't
follow the medication protocol, did not go to the physician when
scheduled, etc. etc. Are the results statistically significant? Was the
study a total waste of time and money and effort?
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.

User: "Gregory L. Hansen"
Title: Re: Teaching physics to biology students	21 Mar 2006 01:56:23 PM

In article <dvp11e$mj4$1@eeyore.INS.cwru.edu>,
Andy Resnick <andy.resnick@op.case.edu> wrote:

Gregory L. Hansen wrote:
I sat in on a clinical presentation for childhood asthma. They started
with, I think, 200 people, 100 in the control, 100 were relocated to
cleaner houses- this was a major study, several million dollars. By the
time the study was over, 2 years and 5 physician visits later, I think
maybe 16 to 20 kids *total* were useable: the rest dropped out, didn't
follow the medication protocol, did not go to the physician when
scheduled, etc. etc. Are the results statistically significant? Was the
study a total waste of time and money and effort?

Wow. The relocated kids would have had to get some pretty big
improvements to get a convincing result out of that.
You'd hope that a project like that can be salvaged by looking at, for
instance, some result versus time before the rigorous medication schedule
was abandoned.
--
"The average person, during a single day, deposits in his or her underwear
an amount of fecal bacteria equal to the weight of a quarter of a peanut."
-- Dr. Robert Buckman, Human Wildlife, p119.
.

User: "Andy Resnick"
Title: Re: Teaching physics to biology students	21 Mar 2006 02:22:47 PM

Gregory L. Hansen wrote:

In article <dvp11e$mj4$1@eeyore.INS.cwru.edu>,
Andy Resnick <andy.resnick@op.case.edu> wrote:

Gregory L. Hansen wrote:

I sat in on a clinical presentation for childhood asthma. They started
with, I think, 200 people, 100 in the control, 100 were relocated to
cleaner houses- this was a major study, several million dollars. By the
time the study was over, 2 years and 5 physician visits later, I think
maybe 16 to 20 kids *total* were useable: the rest dropped out, didn't
follow the medication protocol, did not go to the physician when
scheduled, etc. etc. Are the results statistically significant? Was the
study a total waste of time and money and effort?

My point is that *most* clinical trials follow this pattern- high
dropout rates, small sample size, confounding effects. That's one of
the reasons why drug trials are so long and expensive, why most dietary
studies are... inconclusive?..., and why current medical best practices
tend to rely on what has been done before rather than incorporating new
therapies.
And the joke of all of this, biologists will go on and on about hard
hard *physics* is! I mean, compared to the effort expended in clinical
trial design, how could someone claim Physics I labs are difficult? A
waste of time, maybe, but difficult?
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.

User: ""
Title: Re: Teaching physics to biology students	20 Mar 2006 12:34:46 PM

In article <dvmg12$3rg$1@rainier.uits.indiana.edu>,

(Gregory L. Hansen) writes:

In article <nMlTf.2$25.179@news.uchicago.edu>,
<

> wrote:

In article <dvjro6$6tg$1@rainier.uits.indiana.edu>,

(Gregory L. Hansen) writes:

In article <M%6Tf.45$25.3223@news.uchicago.edu>,
<

> wrote:

In article <20060319111228.B79083@emu.uq.edu.au>, Timo Nieminen
<uqtniemi@mailbox.uq.edu.au> writes:

On Sat, 18 Mar 2006

wrote:

Timo Nieminen <uqtniemi@mailbox.uq.edu.au> writes:

On Sat, 18 Mar 2006

wrote:

That's an interesting and somewhat paradoxical situation that keeps
occuring in human affairs. You would think that, as more resources
become available in some area of activity, people will become more
daring, willing to take bigger risks for the chance to net a big
prize. In fact, beyond some point, quite the opposite occurs and a
conservative, risk averse attitude sets in. It is "we can't risk
failure, thre is too much to lose".

A funny idea to apply to science. If a research project fails, it just
fails to add to the body of knowledge - it can't destroy existing
knowledge. Failure is not loss.

I would say even more, any failed research project is a gain. In the
words of Edison, we learned of yet one more thing that doesn't work.
So it does add something to the body of science (granted, oftentimes
the addition is very slim, but sometimes it isn't).

Assuming that the "failure" is published. If it's just ignored, then
nobody not personally involved will know. It's harder to publish null
results.

Which is a pity because they're valuable. Need I remind that one of
the most famous experiments, Michelson-Morley, was a null result.

That has a lot to do with the implications of the failure. The
Michelson-Morley experiment wasn't supposed to give a null result. If it
had given a positive result it would still have been published, but not
nearly as much attention would have been given to it. But if Muldrew's
wonderful light communication produced a null result, the biological
community would have said "Of course it did, why did he even think to
perform that experiment?" Kuhn might have worked the word "paradigm" in
here a few times.

Years ago the Journal of Parapsychology, and possibly others of the kind,
recognized the "file drawer effect". That is, if a "significant" result
is expected to occur one in twenty times, and a hundred experiments are
performed, you'll get some significant results. If only the successes are
published while the others are locked away in the file drawer, the readers
will find a false level of significance in the published literature as a
whole. And so they'd resolved to publish all results, significant or not.

I wonder how important that is in medical and psychological research.
--

I would say, very, very, very important. Sufficient to totally skew
the result once something becomes fashionable (or unfashionable). It
is even worse when people embark on "meta analyses" (a horrid idea
which should be totally banned) where data from multiple studies is
combined in order to improve statistics. Since said multiple studies
ae only selected from the subset which has been published, if there
was any selection bias to begin with, it'll get greatly enhanced in
such meta analysis.

One thought that passed through my head is that if, say, some drug under
study has such a small effect that it's not obvious on each and every
trial, such that there would be essentially no studies filed away, then
there would be little point in even continuing to develop it.

In the meta analyses, maybe "If the null hypothesis were true then the
number of unpublished experiments expected for this level of significance
is..." would have some meaning.

Yes, right.

I've read a scathing review of obesity research ("The Obesity Myth" by
Campos) where, the author claims, most of it is done by companies that
develop related drugs, half of nutritionists have or have had eating
disorders, and basically nothing claimed about the health hazards is true.
Some authors have even come to conclusions that aren't supported by the
very data they publish.

I haven't studied the literature to the extent that I can say whether I
independently agree or disagree with the author. But from what I have
studied it seems being too skinny is worse than being too fat, and in at
least one study the optimum weight for longevity is higher than you'd get
from the BMI. There was one study where they measured health indicators
before and after liposuction of some women and found no difference at all;
they speculated that the wrong fat was lost, or maybe the fat was lost in
the wrong way, but the possibility that it had nothing to do with fat
didn't seem to occur to the researchers (a notion that Campos accused the
industry of not just rejecting, but not even occuring to them in the first
place so that it could be rejected).

That's often the problem with many statistical studies. Things "just
don't occur" to the researchers since they're not working to find what
is happening. They already "know" what is happening, they're just
working to "prove" it.

My brother has been wanting me to read a book in the same vein about
cholesterol which points out, for instance, the low rates of heart disease
among people like the Masai whose diets have the highest levels of
saturated fats.

Regarding what to make of data, the parapsychologists have tried testing
hypotheses in previously published data. One is that psychic abilities
will be highest at the beginning of a run when interest is higher, wane as
boredom sets in, and return as it nears completion. I think it's
legitimate looking for a pattern like that in data that was taken before
such an hypothesis had occured to the field in general, or in data that
was taken without the intention of testing it. Because filed away or not,
if null then we could expect equal numbers of right-side up and upside
down U's.

Yes, quite right.

I think there must be a lot to learn from parapsychology because more than
any other field they've faced continuing criticism from people convinced
that there was nothing to study and intent on figuring out why they're
wrong. I can only imagine what would result if medical science were held
to the same standards that CSICOP holds the parapsychologists to.

I'm all for it. The number of results from mecical studies which were
subsequently overturned, of "breakthroughs" which subsequently quietly
disappeared, of recommendations which were withdrawn or turned around,
etc, etc., is troubling.
Mati Meron | "When you argue with a fool,
meron@cars.uchicago.edu | chances are he is doing just the same"
.

User: "Timo Nieminen"
Title: Re: Teaching physics to biology students	19 Mar 2006 12:45:56 AM

On Sun, 19 Mar 2006

wrote:

Timo Nieminen <uqtniemi@mailbox.uq.edu.au> writes:

A lot of the possible innovative research isn't that expensive.

Oh, certainly. I was addressing myself specifically to PD"s comment
about HEP experiments. These are very expensive.

HEP _facilities_ are expensive. Are HEP _experiments_ expensive? The
possible (much cheaper) analog in optics would be femtosecond lasers, good
examples of which are expensive ~ $200K, but once you have it, an
experiment only costs the labour and incidentals. OTOH, if the cost of
doing an experiment is high ...
--
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
E-prints: http://eprint.uq.edu.au/view/person/Nieminen,_Timo_A..html
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
.

User: ""
Title: Re: Teaching physics to biology students	20 Mar 2006 02:06:30 PM

In article <dvmsk6$6so$1@eeyore.INS.cwru.edu>, Andy Resnick <andy.resnick@op.case.edu> writes:

Timo Nieminen wrote:

On Sun, 19 Mar 2006

wrote:

Timo Nieminen <uqtniemi@mailbox.uq.edu.au> writes:

A lot of the possible innovative research isn't that expensive.

Oh, certainly. I was addressing myself specifically to PD"s comment
about HEP experiments. These are very expensive.

That's a fallacy- a significant fraction of the cost of science is
people. Training, stipends, etc. etc. Typically, 70-80% of my costs on
a grant are salary: mine, the lab techs, the students, travel, indirect
costs, etc. etc. When all of that is figured in, the cost of materials
and equipment is incidental.

I can provide an example from our beamline. Even during the
construction phase personel expenses amounted to nearly half the
budget. Once major construction was finished, they amount to about
80%.
Mati Meron | "When you argue with a fool,
meron@cars.uchicago.edu | chances are he is doing just the same"
.