A short time ago, I solicited help from listserv members to assist me in
putting together a research methods/stat course for physicians/medical
residents. A receive great help---1 member asked me to post the info
that I received. Here it is:
For some research methods course material, see
http://www.mursuky.edu/polcrjlst/P660f99.htm
Mark J. Wattier, Professor and MPA Director
--------------
You might consider a new edition (4th) of the DeVore & Peck
elementary
stat book, Statistics: the exploration and analysis of data. If your
students use calculators, there is a variation of the 4th edition,
different
title, coming out also. (Disclaimer: I am one of the authors.) Twenty
hours sounds a little short for the book, but (IMHO) it would make a
terrific reference for the students as they learn more statistics.
To get more specific information, contact Roxy Peck at
[EMAIL PROTECTED]
-----------------------
Feel free to draw from my course webpage on
research methodology:
http://www.hs.ttu.edu/Hdfs3390/Default.htm
At first, you'll see the syllabus, but if you scroll
to the very bottom, you can click on any of the
substantive topics, and it will bring up my lecture
overheads on that topic. Links to other websites
related to the topic are also included within each
topical unit.
Alan Reifman
Texas Tech University
-------------------------
I was forwarded your message. I highly recommend a book by edited by
Stephen
Hulley and Steven Cummings called "Designing Clinical Research"
published by
Williams and Wilkins, 1988. It is an inexpensive paperback - very
understandable. I would also recommend that you find out more about the
ORACLE program at UC San Francisco - it sounds similar to what you are
doing. Try going to the www.ucsf.edu web site and looking for ORACLE.
Good
luck.
Cassi Vieten, PhD
Director, UCSF Family Alcoholism Study
------------------------------
I'd suggest my own Design & Stat book from Brooks/Cole except for a
small
problem. The publisher has elected not to publish a new edition, and
what's
left in the warehouse is all there there's going to be. But the good
news
is that there are probably enough copies in the warehouse and in the
(damned) used market to keep a course going for several years.
If you want to see the book, get hold of your B/C rep or check them on
the
web (brookscole.com, I think). They may say "It's out of print," but
that's
incorrect--there are books there in the warehouse. It also doesn't
appear
in B/C's catalog either, adding to the error and misperception. Just
hassle
'em--they WILL process an order.
Even better news--I've located a publisher who wants to take over the
book.
But it'll be a few years before the new edition will be out.
Dick Lehman
-----------------------
I presume you mean seat hours and not credit hours.
One thing to emphasize in such a course is that nobody can a graduate
education
in that amount of time. (I.e., you can't get a doctoral level of
education in 20
hours.) The course can only give them enough background to 1) help them
find
consultants and more 2) work more effectivly with consultants.
*** I would heavily emphasize thinking and concepts. ***
See the archives of edstat-l for long discussions of this topic.
(although
focused on a semester course). You might also post this question there.
You might develop an annotated bibligraphy.
I would take a short time to get down the ideas of a data matrix with
rows
representing cases, columns representing variables, and emphasizing that
measurement is the process of filling in the cells.
I would also emhasize that different many kinds of approaches differ
mainly in
that units make up cases. e.g., a survey may have households,
individuals,
focus groups, couples, platoons, hospitals; an experiment might have
have maybe
mice or people; a content analysis has pictures, x-rays, pieces of text,
video
clips, etc; a meta-analysis has studies. I would emphasize the
importance of
having a grasp of how the cases are selected, how the variables are
selected,
and how the measurement is done.
I would provide a list of the 14 (?) terms in Campbell, D.T. and
Stanley, J.C.
"Quasi-Experimentation" (approximate title)
and the list of sources of bias from Kish, L. ( 1987). Statistical
Design for
Research P. 79.
I would in include the bibliography Huff's How to Lie with Statistcs;
Paulos' a
mathematician reads the news; Abelsons' Statistics as pricipled
argument; list
of the Sage quantitative series; etc.
I would distinguish noise from directional bias; precision from
accuracy;
------------------
Take a look at Harvey Motulsky's book "Intuitive Biostatistics." It's
readable, has good examples, and not so technical as to throw people
off.
While not perfect, it's the best I've seen of its kind. (I understand
Dr.
Motulsky is doing a revision, but I don't know when that might come
out. The
book is published by Oxford University Press, if I remember right.)
Steve Albert
----------------------
Scott,
I suggest the text by Geof and Elizabeth Lofus. Its use of grapths and
confidence intervals to reveal the real meaning of data--instead of
mindless significance tests--makes it ideal.
Frank Schmidt
------
Again, thanks to all for the excellent advice and materials.
SR Millis
--
Scott R. Millis, PhD, ABPP
Kessler Medical Rehabilitation Research & Education Corp
1199 Pleasant Valley Way
West Orange, New Jersey 07052
Tel: 973.243.6976
Fax: 973.243.6990
Emails: [EMAIL PROTECTED]
[EMAIL PROTECTED]
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
