you have it right. ideally, we want a control group which is the same in
all respects other than the treatment variable. if not then you must
statistically adjust in some way.
"variation in the key explanatory variable" is pretty easy to evaluate in
quantitative work, but it is something you have to pay close attention to
in qualitative work (by thinking of the qualitative work as quantitative
and seeing whether there is indeed variation in X, and whether the
controls are adequate).
Gary
On Tue, 15 Apr 2003, Yongwook Ryu wrote:
In class, Gary mentioned the neccessity of having a
control group in
experiements, in order to establish causal relationship between the
explanatory variable and the dependent variable. In regression though, is
this the same as having variation in the explanatory variable? Or is there
difference between the two and, if so, can someone kindly explain it to me?
yongwook
-----------------------------
Yongwook Ryu
PhD Student
Department of Government
Harvard University
Tel:617-493-3397
Email: yryu(a)fas.harvard.edu
-----------------------------
_______________________________________________
gov2001-l mailing list
gov2001-l(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov2001-l