Hi. I have conducted mediation analyses using Mlmed. My within-subjects indirect effect is negative and significant. My within-subjects index of moderated mediation is also negative and significant (confidence intervals do not cross zero). This tells me that including a moderator decreases the indirect effect. My question is this --> If it decreases an already negative indirect effect, does this mean that actually the indirect effect has got larger?
I'm looking at the relationship between OCD and depression over time. The indirect effect is negative (as time increases, OCD decreases which decreases depression). I would expect that with larger values of my moderator, that the indirect effect would get larger.
you can follow Hayes (2017) and add the plugin of PROCESS MACRO for SPSS. this helps you to analyze moderation and in a very smooth way. interaction plots can show you exactly what your moderator is doing. if you already have a negative effect (direct or indirect) and your moderator is further negatively decreasing the effect, then you can say that it suppresses the relationship between or among your other variables.
hence, you can interpret that your moderator further increases the negative effect. in other words, through existence of your moderator, the negative impact is further pushed down. so whatever your moderator is and if it is a "suppressing" variable, it means it further helps to decrease the depression of individuals.
i strongly recommend you draw an interaction plot using means and standard deviation of your DV,, IV and M and see how it interacts with high/low/medium intensity.
if i got your question right, i hope this would help.
Dear Vicky Simkin
When you have Moderated mediation variable, it is good to use Amos, or Smart-PLS, however, regarding your question, when the independent variable indirectly not impacted on dependent variable, or has week impact, it means the there is direct impact of the independent variable on dependent variable.
Best Wishes
you can follow Hayes (2017) and add the plugin of PROCESS MACRO for SPSS. this helps you to analyze moderation and in a very smooth way. interaction plots can show you exactly what your moderator is doing. if you already have a negative effect (direct or indirect) and your moderator is further negatively decreasing the effect, then you can say that it suppresses the relationship between or among your other variables.
hence, you can interpret that your moderator further increases the negative effect. in other words, through existence of your moderator, the negative impact is further pushed down. so whatever your moderator is and if it is a "suppressing" variable, it means it further helps to decrease the depression of individuals.
i strongly recommend you draw an interaction plot using means and standard deviation of your DV,, IV and M and see how it interacts with high/low/medium intensity.
if i got your question right, i hope this would help.
I would suggest that you look at James Gaskin’s videos. They are good for novice in SEM. He also provides nice plugins for AMOS.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics