If teaching about inappropriate causal inferences from correlational
studies is your bent, you might want to integrate some critical thinking
exercises in your lesson plans. I give time in most of my classes to these
issues but found that when students read their texts or assigned articles,
authors frequently imply or use causal language from correlational data. I
am starting to develop this into a regular exercise where students explore a
wide range of material--from magazine articles to jounal articles, where
they might examine key problems in such studies. Students invariably jump
on weak sampling and inappropriate generalization but have a harder time
understanding the internal validity issues of research design. Hence,
_I_harp on random assignments, basic control procedures, and potential
confounding by having the students imagine themselves participating in the
study. I encourage them to explore issues of causality and correlation and
the slippery ways authors use causal language. For example, it might be
useful to see if titles and/or discussion sections are the places authors
make such claims as opposed to more carefully presenting things in the
results section. Are apa journals or different areas more guilty than
others?
Anyway, I am finding this to be a useful assignment, and thought
tipsters might find it useful or have similar ideas for exercises along
these lines. Gary Peterson
Gerald (Gary) L. Peterson, Ph.D.
Professor, Department of Psychology
Saginaw Valley State University
University Center, MI 48710
[EMAIL PROTECTED]
1-517-790-4491
