i take it what you mean is that they compute the se's wrong, you do it
right, when when you do their results go away. in addition to dan's
points, you might try to say something productive at the end, such as more
data are needed to figure out whether the hypotheses are valid. Gary
On Sat, 29 Apr 2006, Dan Hopkins wrote:
Sounds like it will be a cool paper. Here are my
recommendations:
1) The title doesn't give a clear indication of exactly what you
demonstrate--once we have re-evaulated, what conclusion should we draw?
Your title is your first chance to state your contribution, not just your
topic.
You also might be able to 2) consolidate the first two sentences, giving
you more room to focus on your own contribution. Also, 3) you might take
an extra sentence to explain more fully and more generally how exactly you
are contributing--those unfamiliar with the structure of the data might
not understand why the sampling variability is so important. And you
might consider 4) stating exactly what you find in that final sentence--is
it that you find a small/null effect, or that you find the confidence
bounds are too wide to make any concrete statements at all? It sounds
like the latter, but I wasn't sure.
Best,
Dan
On Sat, 29 Apr 2006, Thomas Soehl wrote:
Re-evaluating the effect of bonding social
capital on the quality of
local representation.
Social capital is widely viewed as the cornerstone of functioning
democratic governance. Countering this consensus, Hill and
Matsubayashi (2005) argues that membership in voluntary organizations
which foster 'bonding relations' is negatively associated with the
quality of democratic representation in a community. Our paper
examines these findings by simulating the sampling variability
inherent in the community-level indices that Hill and Matsubayashi
(2005) aggregate from individual-level data. We find that when taking
into account this measurement error of the community level
information, the models do not allow conclusive findings as claimed.
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l