At 03:00 PM 6/9/2009, John Antonakis wrote:
Hi Richard:
By plotting I meant:
1. Estimate a regression model that is testing specific theoretical
propositions
2. Plot the fitted model to help understand the nature of the
interaction (I can work it out from the signs of the coefficients;
however, others may need to see the shape of the data)
That's all I meant; there is nothing data-driven about plotting a fitted
model, right?
Sorry, I understand what you mean now. What I was talking about is
situations where people plot the data, and then decide, say, that an
x^2 term or a log transformation or whatever is called for. You
often hear the suggestion that people plot their data, and there is a
lot of merit to that, but under the conditions I describe I think it
does offer some of the same sorts of problems that sw does.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/
