0

I have one dummy independent variable, one categorical independent variable with six categories, and two continuous independent variables. The dependent variable is binary (yes/no). The interaction term is defined between the dummy variable and one of the continuous variables. Another interaction term I want to examine is between the two continuous variables themselves.

Is it possible to calculate effect sizes in a logistic regression, specifically the effect size of an interaction term in a multiple logistic regression model?

I would greatly appreciate your guidance on how to perform this calculation. I'm currently using SPSS for my analysis.

Additionally, any references you can provide for the calculation method would be very helpful.

Thank you in advance.

MAIMAU
  • 11
  • There are many posts with effect size questions. – user78229 Aug 17 '23 at 11:21
  • As an aside, this is about multiple logistic, not multivariate logistic. The latter term (and the tag) are for when you have more than one dependent variable. – Peter Flom Aug 17 '23 at 11:25
  • Does this answer your question? https://stats.stackexchange.com/questions/621010/determining-the-effect-of-two-coefficients-combined-when-the-model-includes-an-i – Peter Flom Aug 17 '23 at 11:27
  • @PeterFlom Thank you for pointing that out, I amended it. I'm not sure if this is what I'm looking for. I found different ways to calculate effect size but I'm not sure which is the most appropriate approach for a logistic regression with interactions of different types of independent variables. For example, here (https://www.youtube.com/watch?v=W8ktaSKVCL0) they mention cohen's d calculated using that method might not be good for continous iv. – MAIMAU Aug 17 '23 at 16:45

0 Answers0