Renan,
From a completely different direction, I have been teaching an
introductory political behavior class to undergrads, made up of a very
heterogeneous mix of students, for the past four years. The course
includes a statistical literacy component. The syllabus is here:
http://www.reed.edu/~gronkep/formystudents.html, click on any of the
"Pol210" links.
Teaching statistical literacy in a short period is a real challenge.
You're surely going to err on the side of believing that the students
need the theoretical and methodological background necessary to
understand what the statistics mean rather than just giving them useful
guidelines for statistical interpretation. It is a very hard line to walk.
The best book for this, in my opinion, is the small Berry and Sanders
"Understanding Multivariate Research." One step up from that (but a
LONG step) is Weisberg, Krosnick, and Bowen's "An Introduction to Survey
Research, Polling, and Data Analysis." This latter is oriented toward
surveys and public opinion, and may not be applicable. Finally, you
might think about having them buy Johnson, Joslyn, and Reynolds
"Political Science Research Methods" which is far too long for such a
short course, but is a good reference guide with accessible sections on
statistics (as well as a variety of research methods topics).
Now to the course. Let me toot my own horn for a moment and urge you to
consider the model that I've developed for Pol210. Structure the course
so that the students have to get their hands dirty with real data
analysis. I think this is *vital* in really helping students understand
statistics beyond just a very superficial ability to read them. It is
possible to do in a short period of time.
Second, figure out a substantive "hook" for the course. One model is G
King's replication course; another is my own course that focuses on the
paradox of turnout; another in your own field might focus on the
correlates of democraticization or regime stability.
You start with the question / substance, and then walk systematically
through more complex statistical approaches to the problem. At each
phase, the students have to produce a brief memorandum with data
analytic results and an interpretation. Build to a final paper or
project that produces a well-specified multiple regression, logit, or
probit model.
How to deal with the learning curve on the statistical program? Look at
sda.berkeley.edu or the "data access and analysis" system at ICPSR
(http://www.icpsr.umich.edu/access/index.html). This way, you don't
teach them to use Stata, R, or associated--they use the online system.
Renan Levine wrote:
> Dear All,
>
> Does anyone offer a short course on statistics for non-quantitative
> graduate students focused on their ability to read and understand work
> that presents statistics?
>
> I am an assistant professor at the University of Toronto, where not only
> do we not require any quantitative methods courses for our grad
> students, we also have not offered any for a couple of years. We do
> offer a research methods and design course, stats at the undergrad level
> and a qualitative methods class. This is expected to change soon.
>
> In the meantime, some of the grad students have asked the faculty if it
> was possible to offer a short 1 or 2 week statistics class targeted at
> those students with little or no statistics experience. These students
> either do not have sufficient interest to take a semester- long course
> or are already ABD. They envision a course that will be geared towards
> empowering them to better understand the polisci readings that employ
> quantitative methods. So, the class would probably emphasize reading and
> statistical literacy over completing problem sets or running analyses
> and turning in write-ups. Perhaps the class might augment the research
> design course and help them think about how quantitative analysis might
> prove useful in their own research? Cover a few techniques to display or
> analyze data? Call the class "T-tests for theorists" or a stats "boot
> camp" much like other departments' math boot camps (except, skip the
> math!).
>
> If you are aware of any such classes, have a syllabus, or just have a
> few suggestions, please let me know.
>
> Yours,
> Renan Levine
>
> --
> Renan Levine
> Department of Political Science
> University of Toronto
> [log in to unmask]
> http://individual.utoronto.ca/renan
> (416) 978 - 0743
>
> **********************************************************
> Political Methodology E-Mail List
> Editor: Karen Long Jusko <[log in to unmask]>
> **********************************************************
> Send messages to [log in to unmask]
> To join the list, cancel your subscription, or modify
> your subscription settings visit:
>
> http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
--
Paul Gronke Direct phone: 503-517-7393
Dept. of Political Science E-Fax (preferred): 440-274-8159 **NEW**
Reed College Local fax: 503-777-7776
3203 SE Woodstock Blvd
Portland, OR 97202 http://www.reed.edu/~gronkep
**********************************************************
Political Methodology E-Mail List
Editor: Karen Long Jusko <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|
