On Wed, 28 Mar 2001, Harry Avis wrote:
> Good multiple choice
> questions can require fairly high levels of thinking and reasoning. My
> concern about essays and such is reliability. I do not think that anyone can
> be objective in scoring an essay unless they confine themselves to scaning
> for certain key words, which makes them a form of multiple choice. My mood,
> my time constraints, the exam I previously read, the penmanship, my personal
> attitude toward the student, the verbal facility, spelling and grammar are
> all variables that are almost impossible to control for and significantly
> reduce reliability, not to mention validity.
I'm with Harry. I think m-c questions are unfairly maligned. I'm
talking about good questions (the kind I write), not those that
you get in the typical textbook data bank. Since students as well
as faculty buy into the myth, I like to quote the view of the
eminent psychometrist Anne Anastasi at my students, to wit:
"Multiple-choice tests have largely replaced essay questions
in standardized testing programmes, not only because of time
restriction but also--and more important--because they provide
broader subject-matter coverage, yield more reliable and valid
scores, and are fairer to individuals"
-Anne Anastasi, Psychological Testing, 6th ed. 1988
She says more than that, but that gives them the idea. And if
it's good enough for the expert Anne, it's good enough for me.
(by "time restriction" she means that essay questions take so
long to answer you can't test very much of the course in the
typical exam period; with m-p you can test much more)
-Stephen
------------------------------------------------------------------------
Stephen Black, Ph.D. tel: (819) 822-9600 ext 2470
Department of Psychology fax: (819) 822-9661
Bishop's University e-mail: [EMAIL PROTECTED]
Lennoxville, QC
J1M 1Z7
Canada Department web page at http://www.ubishops.ca/ccc/div/soc/psy
Check out TIPS listserv for teachers of psychology at:
http://www.frostburg.edu/dept/psyc/southerly/tips/
------------------------------------------------------------------------
