Moderator is the variable that has an effect on the interrelationship between the two variables which have a zero order correlation.
In another case an importance of the mediator variables has to be seen and it has been known by the researchers that the mediator variables have been seen to be following the models that are given by the researchers (Briggs, 2006). One of the models realizes that interferes between the response as well as the stimulus that an organism has. This model has been seen to represent the complete formulation of the mediation hypotheses that has been recognized by the social sciences researchers. In the general terms it has been said that the mediator can be defined as a variable when the relationship between the predictor and the criterion is defined and is established in an analytical manner. The external events when assume an importance of the psychological significance only than the importance of the mediators is to be realized. In this case the mediator model can be represented by the following diagrams.
The mediator variable can be represented by the path diagram and this model has been seen to assume the three variable systems in which there are two main causal paths that join the variable, the path measures the direct impact of the indirect variable and this also measures the impact of the mediator (Baron, and Kenny, 1986, 1180).
The variable in this case has been seen to function as the mediator when the specific set of characteristics are being followed by the variable and these are inclusive of the variations that might be noticed in the independent variable and these are the changes that have been seen to notify the changes that are related to the mediator. In a similar manner the variations that take place in the mediators notify the changes in the variable (Hayes, and Preacher, 2008, p. 880).
Strength of the relationship between the two variables can be measured by the effect size. Here it can be said that the effect size is the research on a numeric scale. It is the effect size that can be helpful in making the measures if the deference between the two variables is real or if the deference is because of the change that has occurred in the various factor that have caused the changes in the values (Klein, Fan, & Preacher, 2006, p. 100). It has been seen that in case of the hypotheses testing the effect size, the sample size as well as the critical significance levels are the ones that have been given an importance and these are the ones that have been leveled with each other in hypotheses testing.
An importance of the effect size has been determined in the case of the Meta analysis and in this case it has been seen that the effect size is the one that takes into consideration the different studies and it has been seen that in this case the different studies are combined together into the single studies (Briggs, 2006).
It has been said that for the purpose of the Meta analysis the kind of effect size being used is the one that is in correlation with the sum of all the researches being taken into consideration.