I have found that one of the slopes of my moderator variable is not significant alone, but when the slope is combined with the other slope in the entire model, it becomes significant. For instance, when the slopes of HIV-sigma (High and Low) are combined in the model, the overall effect of the moderator variable becomes significant.
Holger Steinmetz
Hi,
yes of course, that's possible. For instance if you have a cross-over interaction where the direction of the effect of one variable completely changes across the moderator range, the "main effect" will be zero.
Best,
Holger
2 votes
2 thanks
Muziwandile Qiniso Luthuli
Dear Dr. Holger Steinmetz
Thank you for your response, it is really helpful.
Regards,
Muziwandile
0 votes
0 thanks
- Badges
- Science topic
Similar questions and discussions