Naturally, what the customers expect/demand/instruct can be a tricky
issue -- not least in medical science, which I guess to be your field.
I had someone come in the door once wanting a factor analysis and going
out with a logit regression, and he was happier than when he came in,
and we got a joint publication out of it, but it isn't always like that.
Nick
[email protected]
Ariel Linden, DrPH
As always, you enlighten and stimulate! I like the concept of Q-Q Plots.
I
typically use them in conjunction with propensity scoring to demonstrate
(visually)that the matches achieved balance on baseline characteristics.
That said, there are some audiences that prefer to see something that
looks
and feels like a histrogram or density curve to see how the
distributions
overlap (or don't). In this situation, I usually offer up a couple
different
methods of visually describing the data.
From: "Nick Cox" <[email protected]>
For comparison of precisely two distributions -- especially when there
is not a prior prejudice or hypothesis that we are expecting
approximations to some named, equation-specified distribution -- I
regard quantile-quantile plots as near optimal. They allow you to focus
on both similarity and differences and to think directly in terms of
what is being measured.
Why is this graph form which is
(a) information-rich
(b) free of arbitrary assumptions (bin or kernel width, etc.)
(c) easy to explain
(d) easy to compute
(e) well documented
not enormously better known?
See -qqplot-.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
