Hey everyone. This is my first post here so please be kind to me
For my thesis my aim is to run a moderation analysis between organizational commitment and training. My dependent variable is job satisfaction.
Org.commitment comes from 4 5-point likert-items, therefore I calculated the median for each observation, in order to get a total measure of org.commitment and to treat this variable as continuous. Training comes from a 6-point likert item. Out of this variable, I created 2 dummy variables: high training and low training.
Therefore, one interaction term is org.commitment * high training.
I suppose that there is no point to include org. commitment *low training if I am not mistaken. The way I think it is that the effect of training is different at different levels of org.commitment. My question is: is it possible to think of org.commitment as the moderator variable? All the examples that I have found online, when interacting a continuous variable with a dummy variable, they treat the dummy variable as the moderator and not the continuous one, as I want to do.
When running OLS analysis, the coefficient that I get is:
• Dummy variable for high training: ,153
• Org. commitment (centered): ,335
• Interaction term: Org.commitment(centered) * high training: -,022
Does it mean that when organizational commitment increases, the effect of high training on job satisfaction decreases? Or in other words, that the positive effect of high training on job satisfaction weakens as org.commitment increases?
When I interact commitment with low training the coefficient is: ,022.
How should I proceed? It is obvious that I am not really good in statistics but I am searching the web literally the whole day but I have not found an answer yet. I anyone can help me I would appreciate it really really a lot! Thanks!!