Good morning,
I have a doubt about a result in a mediation test. I am using Preacher and Hayes process to analyze mediation effects.
My stage is the following:
I have a variable (x) that has a negative effect over the mediating variable (z). When I analyze the effect of X over Y, its effect is positive but not significant, and the problem is when I included the mediation variable (z) in the model, this variable X has a positive and SIGNIFICANT effect over Y. Really I don't know how to explain that... normally the related literature establishes that if there is mediation the significance decreases.
Any sugestion or recommendation???
Thank you in advance.
Kind regards,
Cristina
Hi Cristina Alcaide Muñoz There is a confusion, are you asking about X effect on Y or X effect on Z. What about the Z effect on Y? And the path which become significant when you add Z, is indirect effect of X on Y via Z?
Hi, sorry.
I first test the effect of X and Y, withouth Z, its effect, that is the path c, is NOT significant, but then when I add Z, the effect of X and Y is significant, that is the path c'. I cannot explain that. I found in the related literature, that if there is mediation the significance of c' has to decrease o even desappear.
In this case, the variable X has a negative and significant effect on Z, that is the path a, however the path b, that is Z on Y, is positive and significant.
Thank you in advance.
- Badges
- Science topic
- Similar topics
- Political Science