• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
The Analysis Factor

The Analysis Factor

Statistical Consulting, Resources, and Statistics Workshops for Researchers

  • Home
  • Our Programs
    • Membership
    • Online Workshops
    • Free Webinars
    • Consulting Services
  • About
    • Our Team
    • Our Core Values
    • Our Privacy Policy
    • Employment
    • Collaborate with Us
  • Statistical Resources
  • Contact
  • Blog
  • Login

Exogenous and Endogenous Variables in Structural Equation Modeling

by Karen Grace-Martin Leave a Comment

In most regression models, there is one response variable and one or more predictors. From the model’s point of view, it doesn’t matter if those predictors are there to predict, to moderate, to explain, or to control. All that matters is that they’re all Xs, on the right side of the equation.

Structural equation modeling, though, is a system of equations. Sure, it can be such a simple system that it has only one equation, but it’s flexible enough to have many.

And when you have many equations, the same variable can be a response variable in one equation and a predictor in another. So the simple response/predictor dichotomy breaks down a bit because you may have only a few variables that are only a predictor or only a response.

But there is still an important and related dichotomy: Exogenous and Endogenous. Let’s take a look at what these are.

The difference between exogenous and endogenous

Endogenous variables are response variables in some equation, anywhere in the model. In other words, if somewhere in the model, there is a one-headed arrow pointed to it, the variable is endogenous.

The final response variable, if there is one, is always endogenous. So are mediators. So are any variables we might otherwise think of as predictors, if some other predictor is affecting their value.

In the following diagram, all the x variables are endogenous.

And in this one, we have six endogenous variables: M₁, a₁, a₂, a₃, Y, and M₂.

Exogenous variables are never, ever response variables. They can be correlated with other variables, so have one end of a two-headed arrow pointed toward them. But never a one-headed arrow.

Why this difference matters

When you make a variable exogenous, you’re assuming there are no other variables causing or affecting that variable. This is a particularly important assumption if you’re making any causal inferences. This can be easy to assume for some variables, like treatments with random assignment or some demographic variables, (like age) but something to consider carefully with observed variables.

The big structural difference between exogenous and endogenous variables is endogenous variables have error terms. In other words, we’re stating that two different things affect their value:

  1. Other variables, as represented by the arrows pointing to them.
  2. Random unexplained variance, as represented by the error term.

Exogenous variables don’t have error terms. This means you’re also assuming they’re measured without error. Again, that’s a pretty easy assumption for some variables, like whether someone received the treatment or a placebo. It’s a harder assumption to make for others, like income.

So it’s important for the model, and for you, the modeler, to know whether a variable is exogenous or endogenous. If it’s endogenous, there will be a parameter estimate for the variable’s error variance. If it’s exogenous, there won’t be.

Very often, SEM models are written without these error terms. It’s a shortcut that people use who know there should be an error term there and don’t want to clutter up the diagram.

If you’re still learning SEM, or presenting a model to others who are still learning SEM, I recommend always putting the error terms in. Just for the sake of clarity. If you’re reading someone else’s model, whether that error variance is there or implied, you can always count on the arrows.

3 Overlooked Strengths of Structural Equation Modeling
Confirmatory factor analysis (CFA) and path models make up two core building blocks of SEM. Learn how these help you understand how SEM is used.

Tagged With: endogenous, error term, exogenous, response variable, Structural Equation Modeling

Related Posts

  • The Four Models You Meet in Structural Equation Modeling
  • Member Training: A Guide to Latent Variable Models
  • One of the Many Advantages to Running Confirmatory Factor Analysis with a Structural Equation Model
  • First Steps in Structural Equation Modeling: Confirmatory Factor Analysis

Reader Interactions

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Please note that, due to the large number of comments submitted, any questions on problems related to a personal study/project will not be answered. We suggest joining Statistically Speaking, where you have access to a private forum and more resources 24/7.

Primary Sidebar

This Month’s Statistically Speaking Live Training

  • Member Training: A Gentle Introduction to Bootstrapping

Upcoming Free Webinars

Getting Started with R
3 Overlooked Strengths of Structural Equation Modeling
4 Critical Steps in Building Linear Regression Models

Upcoming Workshops

    No Events

Copyright © 2008–2022 The Analysis Factor, LLC. All rights reserved.
877-272-8096   Contact Us

The Analysis Factor uses cookies to ensure that we give you the best experience of our website. If you continue we assume that you consent to receive cookies on all websites from The Analysis Factor.
Continue Privacy Policy
Privacy & Cookies Policy

Privacy Overview

This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary
Always Enabled
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Non-necessary
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
SAVE & ACCEPT