Effect of stereotypes

Caroline Szymeczek-Seay css at med.unc.edu
Sat Nov 4 12:39:41 EST 2000


Hi Patricia,
The Newsweek article, I believe, (I only skimmed it, and this was
several days ago) is largely based on the work of Claude Steele and
colleagues.  I read a very interesting, and more full, report of the
studies described in Newsweek in an article by Steele, "Thin Ice:
"Stereotype Threat" and Black College STudents" which appeared in the
August 1999 Atlantic Monthly.  Among many conclusions reached in the
article (and I'd urge you to read it) is that one good way to approach
students is to tell students that you are using very high standards and
that based on the work from the students that you've seen thus far, that
you are confident that they can meet those standards.  The authors feel
that this will signal to students that your criticism of their work is
based on *standards* rather than race, and that you do not view them
stereotypically.  (this is my paraphrase of Steele's article).

You've prompted me to find the Atlantic Monthly article again and to
reread it!  Thanks!

caroline  

patricia bowne wrote:
> 
> Has anyone else seen the article in this week's Newsweek,
> which claims that women who were reminded they were female before
> a math test did worse on the test, and that women who had been
> exposed to 'bimbo' stereotypes on the TV showed less interest in
> science/math careers?
> 
> Pretty chilling stuff, though I guess women shouldn't feel
> as bad as old people -- the same article says that old people
> who'd been exposed to the stereotype of decrepitude were less
> likely to accept life-prolonging medical intervention.
> 
> Anyway, it raises the interesting possibility that some of the
> bias that's thought to be inherent in tests on which women
> or minorities score poorly may be not in the test but in the
> stereotype that women and minorities will score
> poorly. One study cited found that women did just as well as
> men on a math test if they were told beforehand that women did
> just as well as men on that particular test.
> 
> Would this be a useful strategy for those of us who are teaching?
> I think we could all (if we make up our own tests) quite
> truthfully tell our classes that our tests were designed so
> that there should be no gender gap in performance on them. Many of
> us could even truthfully tell the classes that the assignments and
> learning activities were designed to eliminate the gender gap.
> Folks teaching more than one section of the same course could
> do a controlled experiment.
> 
> I'd give it a shot myself, except that I work in a women's college.
> 
> Pat







More information about the Womenbio mailing list