Time: Mon Aug 11 20:01:52 1997
	by usr08.primenet.com (8.8.5/8.8.5) with SMTP id RAA16541;
	Mon, 11 Aug 1997 17:54:54 -0700 (MST)
Date: Mon, 11 Aug 1997 17:53:46 -0700
To: adbryan@OnRamp.NET
From: Paul Andrew Mitchell [address in tool bar]
Subject: SLS: "They asked for 20 chronic pot smokers."

When I was in graduate school, I spent a 
lot of time studying research methodology.
The author Campbell is one I remember for
having done the most sophisticated work
on research designs, and why almost all
of them are worthless.  "They asked for 
20 chronic pot smokers" tells me, right
there, that the study is headed for the
trash can, as far as reliability is 
concerned.  This kind of self-selection
bias is sure to doom the statistical
integrity of the research design.  Check into 
Campbell's work, if you want the real story.
I am not 100% sure about his name, however.  
A good sociologist, or social science statistician,
would be able to confirm his name, or not. 
I believe he wrote in the 1960's;  I first
learned of him in 1971, so he had to be
prior to that year.

The most reliable research design is called
a "double blind" design.  This means that
the people doing the observations, and the
subjects being observed, are both in the 
dark about the nature of the experimental design.

"Pygmalion in the Classroom" is a classic for
demonstrating the power of this design.  Grammar
school kids were ramdonly assigned to two different
groups.  In one group, the teachers were presented
with previous academic records which over-stated
the capabilities of the children;  in the other
group, the teachers were presented with previous
records which under-stated their capabilities.

At the end of the academic year, both groups
were administered the same standardized academic
tests.  Children whose teachers expected them to
do better, did better, MUCH better;  children 
whose teachers expected them to do worse, did worse,
MUCH worse.  This experimental design has withstood
every possible attack on its validity.

Here's a keyword on which to search, within the
scope of social science methodological literature:

      "regression discontinuity"

I wish I could tell you more here, but my memory
has faded a little bit.

/s/ Paul Mitchell

At 05:34 PM 8/11/97 -0700, you wrote:
>>I would like to point out they already have a dismal project . They assked 
>>for 20  chronic pot smokers to do study. 
>I think this should produce favorable results on overall driving 
>performance. Anyone that smokes mj on a daily basis has long since 
>overcome any impairment that mj may produce. If memory serves, 
>there are no driving restrictions (on the ones that can drive) on 
>the patients receiving the government's pot. It seems like someone 
>recently posted the something similar regarding marinol.
>>What relevance does that have to medical usres? 
>Well, ya gotta know what warnings to put on the label -- like avoid large
>bowls of chips, don't operate the remote control, etc. :)
>>It seems just another attemt to take away the importance of the medical 
>The Bowman Gray study was announced a few months back. I don't think
>it has much to do with the latest NIH report, but ya never know.
>Alan B.

Paul Andrew Mitchell                 : Counselor at Law, federal witness
B.A., Political Science, UCLA;  M.S., Public Administration, U.C. Irvine

tel:     (520) 320-1514: machine; fax: (520) 320-1256: 24-hour/day-night
email:   [address in tool bar]       : using Eudora Pro 3.0.3 on 586 CPU
website: http://www.supremelaw.com   : visit the Supreme Law Library now
ship to: c/o 2509 N. Campbell, #1776 : this is free speech,  at its best
             Tucson, Arizona state   : state zone,  not the federal zone
             Postal Zone 85719/tdc   : USPS delays first class  w/o this

As agents of the Most High, we came here to establish justice.  We shall
not leave, until our mission is accomplished and justice reigns eternal.
[This text formatted on-screen in Courier 11, non-proportional spacing.]


Return to Table of Contents for

Supreme Law School:   E-mail