Welcome to our first session of our second workshop. This time we'll be talking about conditional and direct effects. And todays specific top off for this session is mediated moderation. Up to now we talked about mediation, parallel mediation, serial mediation, and moderation. This time we'll put all these concepts together and to talk about conditional indirect effects. The agenda for today is the quick review of mediation and moderation, then we'll explain why do we need conditional indirect effects. Why do we need it to combine these things to run one overall motto is that of doing a piecemeal analysis. And the we'll talk about mediated moderation and explain what are these models. Finally, we will go through the theorizing about mediated moderation models. So again this quick review about mediation in our prior workshop we went in depth about what mediation models are. So if you wanted to go through that workshop, feel free to watch them and there are five different sessions that you can go back and gain more knowledge on specific topics about mediation and moderation. At this point, I want to highlight that mediation, when we are running mediation models. We focus on the process on the mechanism through which our independent variable influences our dependent variable. We look for answers of how does the independent variable affect our dependent variable. And then we have a serial mediation models and we also have parallel mediation models. With moderation we are looking at Variables that change their strength or the direction of the causal relationship between the independent variable and the dependent variable. If you remember, we are talking about job meaningfulness and job performance. Yet in this case we are looking at the circumstances in which the independent variable job meaningfulness influences job performance. In this case particularly, we are talking about the effect of job meaningfulness or job performance as a function of organization identification. Okay so why do we need the conditional interact effect? So many studies especially studies before 2012 or before 2010, they conducted mediation and moderation models in different analysis. So that was done in a separate fashion. For example, we have Chen & Aryee, 2007, Liu, Liu, Kwan, & Mao, 2009. But the problem here Is that if we do this piece-meal approach, we are not able to clearly understand the underlying mechanism and contingency factors how they jointly influence all these problem, organizational problems in general. So one approaching that some people did their best or to split sample based on higher levels of moderator or low level of the moderator. One problem that we organizational of scholars may have is sample size. We may not have a big enough sample to split the sample in two different small samples, I would say, but on top of that, if we do this, piece theory splitting the samples approach, we may not understand the whole effect an organizational phenomenon. So we do need to take into consideration mediation and moderation to gather if we wanted to get a more robust, a more precise, accurate, estimate for the relationships between dependent variable and independent variables. So you probably have heard this language. Mediated moderation or moderated mediation. Actually we have both and we'll be talking about both during this sessions, okay. The first one that will be discussed today is Mediated Moderation. Mediated Moderation, so what is mediated moderation. So when we think about mediated moderation models, we are thinking about the moderating effects of our moderator, our W. How is that transmitted through our mediator? So we have a moderation model, and then we look at the mechanism that transmits that explains how our interaction influences our independent variable. One paper that I've been working on for quite some time looks at the relationship between manager and subordinate gender match. And its effects on managerial response to a voice. We argue that effect is moderated by social comparison orientation of the manager. And then we theorize that these interaction influences managerial responses to voice via gratitude, okay. So the difference, the main difference, between mediated moderation models and moderated mediation models is your theory. And how you build your model, how you build your theoretical rationale that leads to your test. So how do we theorize mediated moderation models? The first step Is to conduct a moderation model in which we look at the relationship, the independent variable and the dependent variable. And how that change based on levels of the moderator. If you recall, it's exactly the same test that we did when conducting moderation models. But then you have have the second step, and in the second step you needed to conduct a moderation model. But you are dependent variable now is mediator. So you are looking at the interaction term, effects of the interaction term on the mediator. And finally, you'll theorize about the whole model. So now, you have these effects of the interaction term of the independent variable and your moderator influencing or a dependent variable via a mediator. So that's how we'll theorize that our three steps first the interaction of the independent variable and the moderator on the dependent variable. The second step is, theorizing about the relationship between your independent variable and the moderator, on your mediator. And finally you theorize about the whole model. In this session, we talked about. We gave you a quick review of mediation, moderation, we explained why conditional interact effects model are important and then we explained how to theorize about mediated moderation models.
