A **mediator variable** is the variable that causes mediation in the dependent and the independent variables. In other words, it explains the relationship between the dependent variable and the independent variable. The process of complete mediation is defined as the complete intervention caused by the mediator variable. This results in the initial variable no longer affecting the outcome variable. The process of partial mediation is defined as the partial intervention.

The mediation caused by the mediator variable is developed as a mediation model. This model that develops due to the mediation is a causal model. In other words, this means that the mediator variable has been assumed to cause the affect in the outcome variable and not vice versa. In the field of psychology, the mediator variable explains how the external physical events affect the internal psychological significance.

The mediation caused by the variable cannot be defined statistically. On the contrary, statistics can be utilized to assess an assumed meditational model developed by the mediator variable.

Baron and Kenny have given steps for conducting meditational hypotheses. A variable plays a role on the mediator variable under some specific conditions. The conditions are as follows:

If the change in the level of the independent variable significantly accounts for variation in the other variable, then the variable is considered a mediator variable.

If the change in the other variable significantly accounts for the variation in the dependent variable, then the other variable is considered a mediator variable.

If the other variable strongly dominates the significant relationship between the dependent and the independent variable, then the other variable is termed as a mediator variable. In other words, if the relationship between the dependent and the independent variable no longer exists and their variations are controlled by some other variable, then that variable is termed as the mediator variable.

In general, the mediation model examines the relationship between the dependent variable and the independent variable, the relationship between the independent variable and the mediator variable, and the relationship between the dependent variable and the mediator variable.

If the mediator variable is measured with less than perfect consistency, then the effects caused are likely to be biased. In other words, the effect of the mediator variable is likely to be underestimated and the effect of the independent variable and the independent variable is likely to be overestimated. This bias in the variation of the variable is generally due to measurement error. An Instrumental variable is then used to solve this problem of bias in the variability. If this approach does not work, then the researcher is required to explain that since the reliability of the mediator variable is very high, the bias caused is fairly minimal.

Schedule a time to speak with an expert using the link below.

**User-friendly Software**

Transform raw data to written, interpreted, APA formatted results in seconds.

If the mediation caused by the mediator variable is perfect in nature, then the independent variable and the mediator variable are correlated to each other. This correlation is termed as collinearity. If the independent variable explains all the variation caused by the mediator variable, there will not be any unique variation that would explain the dependent variable, and this will thus result in multicollinearity.

Multicollinearity is generally expected in the mediational analysis of the mediator variable and the dependent and the independent variable, and therefore it cannot be avoided by the researcher.

**Related Pages:**