On 5 Jun 2000 09:13:08 -0700, [EMAIL PROTECTED] (David
R. McCormack) wrote:
> The following is a question posed by a faculty member in the social
> sciences. He has done a logistic regression finding near significance.
> But his Betas are zero. What does this mean? Is it just that the accuracy
> is only to three places? His output table is below.
< snip, example >
The easiest thing it means -- *might* be that he ought to provide his
variables as scores in the 0.1 to 100 range. One reason : so that
coefficients won't be rounded-off to zero. The obvious way to check
is to move the decimal over a few places by using NewX= X/1000 (or
whatever). I hope this was just in the way of confirmation; if that
is the help he needs, then he needs more help than that.
Another check would be to use regression on the 0/1 criterion. OLS
regression has safer programs, surer algorithms, and a wider range of
standard descriptive statistics (for instance: beta and also
standardized beta; means and correlations of the predictors).
If the scores are not huge, then the N is millions or the computer
program is wrong. Or someone is reading the output wrong (most likely
of these choices, if the program is a standard).
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
