Michael aside, I frequently find psych authors themselves using causal-type
language about correlational research. However, I do think we need to help
students explore the issue more rather than just provide the mantra of never
inferring causation from correlation. I am using the Stanovich book in an
entry level scientific foundations class for psych majors. When I cover the
problem of correlation I try to point out the problems (as does Stanovich) of
determining directionality and the possibility third variables. Given this
foundation, I hope students can appreciate how experimental designs attempt to
deal with these issues (but are themselves not perfect). Hence, I prefer
students look not at one study, but understand the value of a larger research
program. In most psych (apa and aps) articles, 2 or more research projects are
usually described.
A case in point is an article by Duckworth and Seligman (2005) titled:
"Self-discipline outdoes IQ in predicting academic performance of adolescents."
It's a longitudinal and correlational approach, where "the effect of
self-discipline on final grades held, even when controlling for first marking
period grades." (from the Abstract, P. 939). The authors conclude, "...that
programs that build self-discipline may be the royal road to building academic
achievement." Wow, now that we have the answer, let's climb down the mountain!
Now, I actually agree with the authors here, because I am somewhat familiar
with the general literature and research as a whole, and of course, the authors
are describing a program of research where possible third variables and
directionality have been explored. I still wish however, that the authors
would be a bit more restrained in using causal language and implying that they
now have found a golden, single, causal, even "royal road" variable! I have a
class where I can try to explain such causal language, but certainly, if I just
simplistically harp about correlation not equalling causation in a Gen. Psych
class, and students see (as they sometimes do) the casual way psych authors
themselves speak about the way one variable "influences, determines, makes, or
produces" a change in behavior from correlational research, then my message is
undermined.
Just some quick thoughts. Now back to spring break. ;-) Gary Peterson
Duckworth, A. L., & Seligman, M. E. P. (2005). Self-discipline outdoes IQ in
predicting academic performance of adolescents. __Psychological
Science,__16(2), 939-944.
Gerald L. (Gary) Peterson, Ph.D.
Professor, Psychology
Saginaw Valley State University
University Center, MI 48710
989-964-4491
[EMAIL PROTECTED]
---
You are currently subscribed to tips as: [email protected]
To unsubscribe send a blank email to [EMAIL PROTECTED]
