Dear All,
I am doing battle at the moment with the following problem and would
appreciate some help!
I have four levels of self-esteem and I can show that participants with
low self esteem are more likely to smoke than those with high levels. I
have done this through chi-square and by plotting some regression lines.
There is a linear relationship as we go through the levels and that is
no problem.
However, I also have six age groups and what I am trying to find out is
if the effect of low self-esteem is more intense on 11-year olds than,
say, 16-year olds. IE: if an 11-year old has low self-esteem are they
more likely to smoke (for whatever reason) than a 16-year old with low
self-esteem?
I have been looking at interactions between self-esteem and age in
Minitab using binary logistic regression, but I am still having
problems. I also have access to SPSS 8 and, for my sins, Statistica
(and a pencil and paper and evern a calculator!)
Any and all help is much appreciated as I've been at this all day.
Many thanks
--
Ian Sutherland
University of Wales
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|