On 6 Mar 2004 08:38:53 -0800, [EMAIL PROTECTED] (mac55) wrote: > Anyone know how to get relative risk in SPSS??? It seems to only want > to spit back an odds ratio. RR is easy enough to do by hand, but it > would be nice and would save me from having to calculate the CI.
The actual 'RR' is only rarely desirable, because it is mathematically intractable and (for most applications) logically indefensible. If it is going to be large enough so that it differs from the OR, so that you need it in particular, you will probably need specialty-software, which starts out by labeling your groups with total explicitness; instead of something written for general applications like SPSS. Questions on SPSS will do better at the SPSS group, comp.soft-sys.stat.spss . -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html - I need a new job, after March 31. Openings? - . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
