← All Posts /


Ever heard things like “number of astronauts dying in space is directly correlated with no. of people not wearing a seatbelt in cars” or “no. of students against DU conducting online exams is directly correlated with no. of people using Netflix”, you may find these things funny and hilarious but there exists a statistical concept related with it. It is known as Spurious Correlation, let’s find out what interesting things this concept has for us in the box.

Spurious Correlation Correlation is used to test the relationship between 2 or more variables. For instance, Weight and height are correlated. A taller person would tend to have more weight than a shorter person. However, this may not be always true. Hence, we use the Correlation coefficient, which measures as to how strong/ weak, the relation is between the variables. The range of correlation coefficient varies between (-1,1). The higher the correlation coefficient, the higher is the positive correlation between the variables and vice versa.


Simply, the importance of correlation in business decision- making means better weighing of factors that have an effect on its performance.

For instance, given are some common business correlations -

Allocating more resources on Research & Development (R&D) correlates with more innovation. The hiring of employees with certain personality characteristics correlates with higher productivity. Correlation is greatly used in predicting the future of a business direction using Regression. If marketers identify a correlation between consumer behavior and events and a particular type of product or service, then they can take advantage of the relationship to boost business.

If measures are taken to correlate unknown factors with business performance, then it may lead to less uncertainty in the Business since these unknown factors may have a huge impact but sometimes, these factors are volatile, complex, unknown; hence it is difficult to take them into consideration.

Correlation is greatly used in various business industries, one such being Marketing Analytics. Managers lookout for the following correlations -

Correlations between on-page keyword use and rankings. Correlation with the type of top-level domain (.com, .org, etc.) and rankings in Google search. Relation of posts with images getting more shares across social media.

Above is a snapshot of an article in The Washington Post about the relationship between the gender of hurricanes’ names and the number of deaths the hurricane causes.

The article’s title is “Female-named hurricanes kill more than male hurricanes because people don’t respect them, study finds.” The author concluded that the number of hurricane-related deaths is caused due to the gender of the hurricane’s name. Now, there may be (read ‘is’) relation between these two variables, however, there is no reason to believe that due to specific naming of the hurricane, it leads to more deaths, since ‘people take feminine hurricane names less seriously than their male counterparts’ and hence, do not prepare themselves as they should.

The author realized the mistake and hence, the title was changed.

Correlation Vs Causation

A basic fact with respect to both these terms is that - “Correlation doesn’t imply Causation”. Let’s understand what this statement means;

Causation, also referred to as ‘ Cause and effect’, is just an extension of Correlation, which says that a change in one variable will cause a change in the value of another value.

Hence, correlation and causation must not be mixed up.

Since now we understand the difference between Correlation and Causation, Let’s move to Spurious Correlation.

SPURIOUS CORRELATION is a relationship wherein two events or variables are associated or correlated with each other, but not causally related i.e. they have no relation or meaning between them. A spurious correlation is usually caused due to coincidence or a third factor that may be ignored at the time of examination, usually known as the ‘lurking variable’ or ‘confounding factor’.

Whenever two events are related in the same direction, we say that there exists a correlation between them but most of the events are spuriously related as we do not find any cause of this relation.

Given is an interesting example of spurious correlation, you may notice that the graph is perfectly correlated but there is no reason or cause between this way- apart- from each other events

Spurious Correlation

Photo credit - http -//www.tylervigen.com

Spurious Correlation In the above example, Ageing is a confounding variable. The apparent association between living in old age homes and having Alzheimer’s is confounded by age.

If a researcher doesn’t take age into consideration, then he/ she may draw incorrect conclusions about the correlation between living in an old age home and Alzheimer’s.

Misrepresentation of Data to show correlation -

Skewed Scales Manipulating Ranges to Align Data

Even when Y axes measure the same category, changing the scales can alter the lines to suggest a correlation. These Y axes for a company’s yearly revenue from a particular country, differ in range and proportional increase. Spurious Correlation

Eliminating the second axis shows how skewed this chart is. Such problems can arise during the analysis due to the misrepresentation of the data. Spurious Correlation

How to Spot Spurious Correlations -

It is very important to spot a Spurious correlation. Some of the methods that can be used are as follows -

•Using a null hypothesis and checking for a high p-value.

•By ensuring there is a proper representative sample.

•By having an adequate sample size.

Well, the truth is that Spurious Correlations are everywhere, you will find it in the news, websites, blog posts, etc. Many times, these correlations may even be used to share fake news.

So, what are you waiting for start googling up and find correlation between events that may amuse you and us!

Recommended » Benford's Law