Hi Holger,
I'm "cc"ing the list since that's an entirely reasonable pair of
questions. Yes, we mean comparing the world where we see these same
observations with all treatment indicators set to 0 to the world where all
treatment indicators are set to 1. And by original sample, we mean three
separate samples: the mail sample (assigned to treatment of one mailing
and controls), the door sample (assigned to treatment of one door-knock
and controls), and the phone sample (assigned to one phone call vs.
controls).
Best,
Dan
----
Ph.D. Student
Department of Government
Harvard University
Tutor, Currier House
dhopkins(a)fas.harvard.edu
http://www.danhopkins.org
On Mon, 17 Apr 2006, Holger Lutz Kern wrote:
Hi Dan and Ian,
I have a question about ps 7 (a). I
estimated the ITT for each of the three
treatments, but am having trouble
understanding what you want us to do
next. Specifically, what do you mean by
the first difference in this case?
Comparing a world in which no one was
assigned to the treatment to a world in
which everyone was? And does whole
sample refer to the original sample with
multiple treatments or the "door",
"mail", "phone" subsamples that we
constructed?
best,
Holger
--
Holger Lutz Kern
Graduate Student
Department of Government
Cornell University
Institute for Quantitative Social Science
Harvard University
1737 Cambridge Street N350
Cambridge, MA 02138
www.people.cornell.edu/pages/hlk23