Distinction between Mediator and Moderator Variable

Mediator variable is the middle variable / "middleman" between an independent variable (IV) and a
dependent variable (DV). Objective of the mediator variable is to explain the relationship between IV
& DV e.g. IV is not directly influencing DV but rather IV is indirectly influencing DV through mediator
variable. Pictorially, Independent variable --> Mediator variable --> Dependent variable. For
example, salary (IV) is positively influencing education (mediator variable) and then education is
positively influencing health-screening expenses (DV). When the effect of education is removed, the
relationship between salary and health-screening disappears.
Moderator variable is a third party variable that modify the relationship between an independent
variable (IV) and a dependent variable (DV). Objective of the moderator variable is to measure the
strength of the relationship between the IV & DV. Pictorially, moderator variable's arrow line is
pointing to the mid point of the arrow-lined relationship between independent variable --> dependent
variable. For example, if age is a moderator variable between salary (IV) and health-screening
expenses (DV), then relationship between salary & health-screening can be stronger for older men
and less strong for younger men.

Mediator vs moderator variables

Published on March 1, 2021 by Pritha Bhandari. Revised on September 15, 2021.

A mediating variable (or mediator) explains the process through which

two variables are related, while a moderating variable (or moderator) affects the
strength and direction of that relationship.

Including mediators and moderators in your research helps you go beyond studying a
simple relationship between two variables for a fuller picture of the real world. These
variables are important to consider when studying complex correlational or causal
relationships between variables.

Table of contents

1.
2.
3.
4.

What’s the difference?

You can think of a mediator as a go-between for two variables. For example, sleep
quality (an independent variable) can affect academic achievement (a dependent
variable) through the mediator of alertness. In a mediation relationship, you can draw an
arrow from an independent variable to a mediator and then from the mediator to the
dependent variable.

In contrast, a moderator is something that acts upon the relationship between two
variables and changes its direction or strength. For example, mental health status may
moderate the relationship between sleep quality and academic achievement: the
relationship might be stronger for people without diagnosed mental health conditions
than for people with them.

In a moderation relationship, you can draw an arrow from the moderator to the
relationship between an independent and dependent variable.
Mediating variables
A mediator is a way in which an independent variable impacts a dependent variable.
It’s part of the causal pathway of an effect, and it tells you how or why an effect takes
place.

If something is a mediator:

1. It’s caused by the independent variable.

2. It influences the dependent variable
3. When it’s taken into account, the statistical correlation between the independent
and dependent variables is higher than when it isn’t considered.

Mediation analysis is a way of statistically testing whether a variable is a mediator

using linear regression analyses or ANOVAs.

In full mediation, a mediator fully explains the relationship between the independent
and dependent variable: without the mediator in the model, there is no relationship.

In partial mediation, there is still a statistical relationship between the independent and
dependent variable even when the mediator is taken out of a model: the mediator only
partially explains the relationship.

Example: Mediator variablesIn a study on socioeconomic status and reading ability in children,
you hypothesize that parental education level is a mediator. This means that socioeconomic
status affects reading ability mainly through its influence on parental education levels.
You use a descriptive research design for this study. After collecting data on each of these
variables, you perform statistical analysis to check whether:

1. Socioeconomic status predicts parental education levels,

2. Parental education levels predicts child reading ability,
3. The correlation between socioeconomic status and child reading ability is greater when
parental education levels are taken into account in your model.
Moderating variables
A moderator influences the level, direction, or presence of a relationship between
variables. It shows you for whom, when, or under what circumstances a relationship will
hold.

Moderators usually help you judge the external validity of your study by identifying the
limitations of when the relationship between variables holds. For example, while social
media use can predict levels of loneliness, this relationship may be stronger for
adolescents than for older adults. Age is a moderator here.

Moderators can be:

 Categorical variables such as ethnicity, race, religion, favorite colors, health

status, or stimulus type,
 Quantitative variables such as age, weight, height, income, or visual stimulus
size.

Example: Moderator variablesIn a study on work experience and salary, you hypothesize that:

 years of work experience predicts salary, when controlling for relevant variables,
 gender identity moderates the relationship between work experience and salary.

This means that the relationship between years of experience and salary would differ between
men, women, and those who do not identify as men or women.
To test this statistically, you perform a multiple regression analysis for the data on work
experience and salary, with gender identity added in the model. You compare the statistical
significance of the model with and without gender identity included to determine whether it
moderates the relationship between work experience and salary.
Moderator Variable
A moderator variable, commonly denoted as just M, is a third variable
that affects the strength of the relationship between a dependent and
independent variable In correlation, a moderator is a third variable
that affects the correlation of two variables. In a causal relationship,
if x is the predictor variable and y is an outcome variable, then z is the
moderator variable that affects the casual relationship of x and y.
Most of the moderator variables measure causal relationship using
regression coefficient. The moderator variable, if found to be
significant, can cause an amplifying or weakening effect between x
and y. In ANOVA, the moderator variable effect is represented by the
interaction effect between the dependent variable and the factor
variable.

Questions Answered:

Does gender effectively moderate the relationship between desire to

marry and attitudes of marriage?

Does Z treatment effect the impact of X drug onto Y symptoms?

Moderated regression analysis: This is a regression based technique

that is used to identify the moderator variable. To explain how MRA
technique works, we can use the following example:

Let
(1)
(2)
(3)

In this equation, if (the interaction between the independent variable

and moderator variable) is not statistically significant, then Z is not a
moderator variable, it is just an independent variable. If is
statistically significant, then Z will be a moderator variable, and thus
moderation is supported.

Linear vs. non-linear measurement: In a regression equation, when the

relationship between the dependent variable and the independent
variable is linear, then the dependent variable may change when the
value of the moderator variable changes. In a linear relationship, the
following equation is used to represent the effect:

In this equation, the relationship is linear and represents the

interaction effect of the moderator and the independent variable.
When the relationship is non-linear, the following equation shows the
effect of the moderator variable effect:

In this equation, the relationship between the dependent and the

independent variable is non-linear, so and shows the interaction effect.
In a repeated measure design moderator, the variable can also be
used. In multi-level modeling, if a variable predicts the effect size,
that variable is called the moderator variable.

1. Alternative: In a non-linear relationship, a significant value of a

moderator variable does not prove the true moderator effect.
Unless the moderator is a manipulated variable, we cannot say if
the moderator variable is a true moderator or if it is just used as
a proxy.
2. Level of measurement: The moderator variable is an
independent variable that is used to measure the causal
relationship. Like other independent variables, it may be
categorized or continuous.

Assumptions:

1. Causal assumption: When x variable is not randomized, then

causation must be assumed. The moderator variable can
 reversely effect the causation, if the causation between x and y
is not presumed.
2. Causal variable relationship: The moderator variable and
independent variable, in principal, should not be related. No
special interpretation can be found between a correlated
independent and moderator variable. However, they should not
be too highly correlated, otherwise, estimation problems may
occur. The moderator variable must be related to the dependent
variable.
3. Measurement: Usually, the moderation effect is represented by
the interaction effect between the the dependent and
independent varaible. In a multiple regression equation, the
moderator variable is as follows:

In this equation, the interaction effect between X and Z measures the

moderation effect. Typically, if there is no significant relationship on
the dependent variable from the interaction between the moderator
and independent variable, moderation is not supported.

Related Pages:

 Correlation
 ANOVA
 Multiple Regression

Statistics Solutions can assist with your quantitative analysis by

assisting you to develop your methodology and results chapters. The
services that we offer include:

Data Analysis Plan

Edit your research questions and null/alternative hypotheses

Write your data analysis plan; specify specific statistics to address the
research questions, the assumptions of the statistics, and justify why
they are the appropriate statistics; provide references
Justify your sample size/power analysis, provide references

Explain your data analysis plan to you so you are comfortable and
confident

Two hours of additional support with your statistician

Quantitative Results Section (Descriptive Statistics, Bivariate and

Multivariate Analyses, Structural Equation Modeling, Path analysis,
HLM, Cluster Analysis)

Clean and code dataset

Conduct descriptive statistics (i.e., mean, standard deviation,

frequency and percent, as appropriate)

Conduct analyses to examine each of your research questions

Write-up results

Provide APA 6th edition tables and figures

Explain chapter 4 findings

Ongoing support for entire results chapter statistics