logistic regression interaction term - change between main effects and full model

#1
Hello,

I would appreciate your help with this problem. I have a binary logistic regression model. The main effects model includes female (coded 0 = male, 1 = female), IV of interest (ranges 0-3) and other controls. The full model includes all of the above plus the interaction of female and the IV of interest.

In the main effects model, both female and the IV of interest have odds ratios above one; female is not significant but the IV is significant. In the full model, the interaction term is significant and the odds ratio above one, the IV also is significant and has an odds ratio above one, but female remains insignificant and has an odds ratio below one.

Does the fact that female goes from being above one in the main effects model to below one in the full model mean anything? Does it affect the interpretation of the female*IV interaction term?

Thank you SO much for any thoughts!

Tucker