Ecologic Studies

["R. William Field" <bill-field@uiowa.edu>]

 R. William Field wrote:

The ecologic study is not more powerful because of the non
quantifiable biases within the data set, you have no idea what types of
cross-level biases have been introduced.  Again, what Dr. Cohen has is a
large data set, this does not make a study more powerful.


Cohen responded 	--If this is the problem you imply that it is, surely you
can make
up a not implausible example. In the 3 years since my paper was published,
and despite the large monetary rewards offered, no one has done this. Your
example does not have to be true, as long as it is not implausible. It
seems to me that the time has come for those who cry "cross level bias" to
put up or shut up.

FIELD RESPONSE ______________________


"put up or shut up"??   This is scientific dialogue??

Dr. Cohen, I believe several individuals have provided you with very
plausible explanations for your findings.

Greenland and Robins (1994)  Invited commentary: ecologic studies - biases,
misconceptions, and counterexamples.  Am. J. Epidemiology  139:747-760; 1994.

in addition to the article in press which you know about. 

I find their explanations very reasonable.

Bill Field
bill-field@uiowa.edu