Correlation vs Causation2

What is Causation?

SHARE THE ARTICLE ON

Table of Contents

What is Causation?

Causation implies a cause-and-effect relationship between variables; a change in one variable ‘causes’ a change in the other. When two variables have a causal relationship, a change in the independent variable (or the ‘cause’) influences a change in the dependent variable (or the ‘effect’). 

There are four main criteria for causality: 

  1. Covariation: The variables must vary together.
  2. Nonspuriousness: The association between the variables must not be due to a third, or confounding, variable. 
  3. Plausibility: The claim of the causation must be plausible, and must therefore make sense. 
  4. Temporality: The “cause” must take place before the “effect”. 

Transform your insight generation process

Create an actionable feedback collection process.

online survey

Deriving Causal Relationships

Let’s take a look at how cause-and-effect relationships are derived in fields of study such as statistics and medicine: 

  • Experimental Data: The most effective way to derive a cause-and-effect relationship is through the use of randomized experimental data. In this method, random assignment is used to separate subjects into a control group and a treatment group to study the effects of the manipulation of the independent variable (or the treatment) by comparing both groups. 
  • Observational Data: In cases where randomized experimental data is unethical or impossible to obtain, observational data is used. Such dara is generally extracted from existing data sources due to which there is an absence of random assignment and control. It is harder to conclusively derive a causal relationship between variables using observational data as the variables can’t be controlled or assigned into different groups through randomization.

Download Market Research Toolkit

Get market research trends guide, Online Surveys guide, Agile Market Research Guide & 5 Market research Template

Making the most of your B2B market research in 2021 PDF 3 s 1.png

Examples of Causation

The following are a few examples of causal relationships between variables: 

  • Price and Demand (The Law of Demand): In the study of economics, the law of demand explains that demand and price have a negative causal relationship where ‘price’ is the independent variable and ‘demand’ is the dependent variable; when price increases, it influences a decrease in demand and when price decreases, it influences an increase in demand. 
  • Caloric Intake and Weight: In this relationship caloric intake is the independent variable and weight is the dependent variable; an increase in a person’s caloric intake leads to an increase in their weight while a decrease in their caloric intake leads to a decrease in their weight. These two variables have a positive causal relationship as both variables move in the same direction. 

See Voxco survey software in action with a Free demo.

The Difference between Causation and Correlation

Correlation is a term that simply implies the existence of a relationship, or statistical association, between variables. Although causation also implies a relationship between variables, it goes farther to define the type of relationship as well. Causal relationships exist only when one variable ‘causes’ a change in the other. 

It is important to note that although causation always implies correlation, correlation does not imply causation. But why?  

For two variables to have a causal relationship, they must have a statistical association, and they must therefore be correlated. However, there are circumstances where correlation does not imply causation. These are a few reasons why correlation does not imply causation: 

  • Sometimes even when variables do have a cause-and-effect relationship, there may be directionality issues. Directionality Issues make it hard to determine which variable is the independent variable and which is the dependent variable, making it impossible to conclusively derive causation.
  • Another reason why correlation does not imply causation is due to the influence of extraneous/confounding variables. In many cases, although it may seem like two variables are very closely related, they may actually be correlated due to a third variable (or a chain reaction of multiple other variables) that has a causal relationship with each one of the variables individually. Therefore, these variables do not have a causal relationship as their statistical association can be due to the influence of another variable. 

FAQs on Causation

Causation implies the existence of a cause-and-effect relationship between variables; when one changes, it causes a change in the other.

Correlation simply implies the existence of a relationship between variables while causation implies the existence of a causal (cause-and-effect) relationship between the variables. Causal relationships exist when a change in one variable influences a change in the other.

 Correlation does not imply causation because; 

  • Directionality issues sometimes make it hard, or impossible, to determine which is the independent variable and which is the dependent variable. 
  • The existence of confounding variables may be the actual reason for the association between the two variables. 

When two variables have a cause-and-effect relationship, the “cause” is known as the independent variable while the “effect” is known as the dependent variable; when there is a change in the independent variable, it causes a change in the dependent variable.

Explore all the survey question types
possible on Voxco

Read more

Hindol Basu 
GM, Voxco Intelligence

Webinar

How to Derive the ROI of a Customer Churn Model

30th November
11:00 AM ET