For this session, we'll be testing a model in which we argue, we theorize that the relationship between meaningfulness and performance is a function of organizational identification. Perhaps, if you have high levels of organization identification, If you highly identified with organization, the relationship between job meaningfulness and performance would be stronger. So if you have low levels of job identification, perhaps the relationship between meaningfulness and performance is low or weaker than when you have high levels of organization identification. We are adopting SPSS and the macro developed by Hayes, the PROCESS macro. So, if you click on Analyze, go to Regression, among all these options that you have here, go to Regression. And then when you click on Regression, you have the PROCESS option. You are not doing a linear regression model, you are not doing ergonomic regression model, you are using the PROCESS macro developed by Hayes. What's important, again, and you should know that now, is the choice of the model. For moderators, for moderating models, we are adopting number one, model number one. Not model number four from mediation, or not number six for serial mediation. Now, we're adopting number one. Our dependent variable is our performance measure. Our independent variable is job meaningfulness. And our moderator is organization identification. You'll notice that we added the moderator in the same place that we added our mediator, but SPSS will know that this is a moderator because we chose model number one. The variables down here propose moderator W, or Z, or V, or Q. They are used when we are conducting conditional indirect effects. And this is the topic of our next workshop. We'll be talking about conditional indirect effects, and you will learn to use those options here in our next workshop. So, what you do next to test from moderator for moderating models? Next, click on Options, remember we needed to mean center our variables. And SPSS PROCESS model does that for you, which is fantastic, helps us or makes these tasks more easy, one last step we needed to do by hand. So, just click on that Mean center for products. And then I strongly recommend you to also select, Print model coefficient covariance matrix, because we'll need that matrix to do the simple slope analysis. And there we go, if we do that, click on OK, and then you have the output file. Again, double check if you got the right model, model number 1. Yay, we can move, and then we have the variables here, perfect, they are all there. Let's look at the coefficients now. So, in our moderation models, we needed to look at the facts of the interaction term. In this case, we have an interaction term that is significant. So the relationship between the interaction term and our dependent variable here performance is significant, p is less than 0.05. And the second thing that you look at your output file is a covariance table. Yeah, you do have the covariance table here. Remember that the covariance of the covariance is the variance. Keep in mind that language because we'll need those coefficients when plotting our interaction graph, and getting the simple slope analysis. How do we know that this is the right interaction, right multiplication term? We look here, you see, so here our int_1, interaction 1 is the multiplication of job meaningfulness and organization identification. Finally, you can look at the change in R-square when we add the interaction term. And here, we see that the change in R-square explained various is significant, and that's important as well. You needed to report that when you are reporting the findings off a moderation model. Finally, the last part of the output that you look at is this set of variables, or set of coefficients that show you, or the set shows you that the relationship is significant for high or low levels of the moderator. At this point here, we look at the mean levels of the moderator, and we find that at mean levels of the moderator, 0 is in the confidence interval, so that effect is not significant. But when you're high in organization identification, we find that the direct effect of job meaningfulness on job performance is significant, there is no 0 there. But, if you are low in organization identification, 0 is in the confidence interval. So, you don't find a significant direct effect of job meaningfulness on performance for those individuals who are low on organization identification. This is the last part of the output. I usually don't use them this coefficient, I prefer going to the Excel spreadsheet because it will give you a more, I would say, an easier way to plot your interactions.