How to do post hoc tests in logistic regression?

#1
I looked at whether participants recycled after being subjected to a given message. There were three groups: a Control group, group A, and group B.

Since my dependent measure was dichotomous (recycling vs. using the waste bin), I performed a logistic regression in SPSS. However, SPSS only let me compare individual groups to the Control group. I found that group A recycled more than controls, but group B did not.

What I am curious about now is whether group A recycled significantly more than group B.

Is it possible to do such a post-hoc test? If so, how?


P.S. Book recommendations are also much appreciated. Currently I'm using Andy Field's Discovering Statistics, but the logistic regression chapter doesn't cover this question.
 

Mean Joe

TS Contributor
#2
I don't use SPSS, so I don't know any syntax.

SPSS only let me compare individual groups to the Control group
There must be some way to change the reference group. From my Googling, it seems SPSS makes the last (alphabetically-speaking) group the reference, so what people do is alter the name of group "A" to something like "ZA".
 
#3
It is certainly possible to change the reference group. I had considered this, but it seems cumbersome in comparison to typical post-hoc tests. Are there any reported studies that do this?

Assuming I were to try it by making A the reference group, would I then need to adjust the alpha value, such as with a Bonferroni correction?