Dear Arun,

I have done this in R. It's easier to find chi square for linear trend.

You can find R in here.

http://cran.r-project.org/bin/windows/base/

Arun, when you do Chi square for linear trend, it means there are more than two groups, usually each group indicating a higher/ increasing exposure. The exposure code means the code given to the group where the first group would be the reference group.

Say for example exposure to cigarette smoke

Group 1= Non smoker = which you can give exposure code 1 (reference group)

Group 2= smoker (10 Ciggs/day) = which you can give exposure code 2

Group 2= smoker (20 Ciggs/day) = which you can give exposure code 3

Thank you sir, I got it.