Welcome to our second session of our second workshop on conditional interactive facts. This time we'll be talking about mediated moderation and we'll show you how to conduct an analysis on mediated moderation models. In the prior session, we went through the theory behind mediated moderation models. Now we'll get our data and run some analysis. So this is the agenda for today. We'll be testing for mediated moderation models. And we'll go through all these steps as we progress. So the first step is actually conducting a traditional test of moderation on the DV. The second step is conducting a traditional moderation test on the mediator. So the only difference is that we are changing our dependent variable instead of using our distal dependent variable, we are using the mediator as our dependent variable in our Step 2. And then in Step 3, we will conduct and analysis using process and specifically the model 8 on the PROCESS macro developed by Hayes. If you don't have that PROCESS macro installed on your computer, this is a good time to install it. So you can download this PROCESS macro and then install on your computer. We need that macro to run this analysis. And then, we'll conduct a bootstrapping analysis, that's our Step 4. And in Step 5, we'll look at the index of moderated mediation. That index is something new that Hayes created for us, so there is a paper and you can download this paper by clicking on the link in the description of this video. So feel free to go there and download that paper if you wanted to know more about the index of moderated mediation. So let's do it. Let's run an analysis. And this is the model we'll be testing today. We'll keep it constant with our first workshop, so we'll be using job meaningfulness in the job performance. And what we are theorizing here is that meaningfulness influences job performance, and that relationship is a function of the organization identification of the employee. And then we theorize that this interaction term is explained, the effects of this interaction term is explained by job engagement. Let's do it. So the first step is testing for an interaction term and the effects of that interaction on job performance. Again, this is a very similar, it's a simple test and you've done this in one of our prior sessions in which we walked through the moderation models. So if you are not familiar, if you need more information on moderations, go to that particular session in our Workshop 1. So here we'll go to Analyze > Regression and again click on the PROCESS macro by Hayes. Enter the information on this screen. And we have as our dependent variable, performance. As independent variable, meaningfulness. As our moderator, organization identification. Here we needed to select the correct model, which is model number 1. Go to Options and select the Mean center for products, that's important to avoid multicollinearity issues. And also print the covariance matrix. Then click on OK. So this is information that you get from that analysis. And remember, we'll be looking at the relationship of the interaction term and our dependent variable performance in this case. We find that the interaction term is significant. And then you'll scroll down the output file and you'll see that for high levels of organization identification, the direct effect of job meaningfulness on job performance is significant. But for low levels of organization identification? No, there is 0 in the confidence interval. And then, we are ready for a Step 2. And for a Step 2, we conduct the same moderation test, but now instead of using job performance as an outcome, we use job engagement, our mediator. You run the same test, go to PROCESS, enter all the variables. But again, instead of using job performance our dependent variable, as our outcome, we use our mediator, job engagement. Don't forget to select model number one to conduct moderation analysis. The same set of outputs that you want. Important, again, to select Mean center for products and to print the covariance matrix. Click on OK.