• Articles
  • Tutorials
  • Interview Questions

What Is Correlation In Statistics? Types and Examples

What Is Correlation In Statistics? Types and Examples

Correlation is a statistical term that is widely used in the field of data analytics. In this blog, we will see what correlation actually means in statistics, its various types, and real-life examples of correlation.

Watch this Data Analytics Course video to learn more about its concepts:

Correlation in Statistics

Let’s answer “what is correlation?”.Correlation in statistics is a type of concept that assesses the strength of an association or relationship between two variables. In other words, correlation is a measurement of the relationship between changes in one variable and changes in another. It is a frequently used tool in data analysis and helps researchers understand the interrelationships between various parameters.

In data analysis, correlation is a potent technique that helps in understanding the intricate connections between various variables. Researchers can create more accurate predictions and draw more well-informed inferences from data by understanding correlation.

Want to know more about data analytics? Enroll in this professional Data Analytics Certification to learn from experts.

Correlation Coefficient

The correlation coefficient is a statistical measure that quantifies the strength and direction of the linear relationship between two variables. 

To define it in a comprehensive way:

  • The correlation coefficient, represented by the letter “r,” is a numerical value that falls within the range of -1 to +1.
  • A negative correlation is indicated when the value of r is less than 0, which is r < 0. 
  • While a positive correlation is indicated when the value of r is greater than 0, which is r > 0
  • A value of r = 0 suggests no correlation between the variables being analyzed. 

Thus, to calculate the correlation coefficient, data is typically inputted into a calculator, computer, or statistics program, as the process can be time-consuming.

Need of Correlation in Statistics

We’ll look at a number of factors that will explain why correlation is a crucial tool:

  • Relationships are Identified: Correlation in statistics helps in the identification of relationships between variables. Researchers can ascertain whether two variables are positively, adversely, or not connected by examining the correlation coefficient. Making forecasts and reaching conclusions based on facts can both benefit from knowing this knowledge.
  • Measure the Strength of the Relationship: Correlation is a tool used by researchers to determine the degree of relationship between two variables. Strong correlations reflect a strong relationship between two variables, whereas weak correlations suggest a weak relationship between two variables.
  • Predictive Modeling: Correlation in statistics allows the development of predictive models. Researchers can use correlation, for instance, to better understand how the prices of various assets change in relation to one another in the field of finance. This understanding helps researchers make more informed investment decisions.
  • Validity of Data: Researchers can assess the validity of data using correlation. The reliability of the data and how precisely it depicts the relationship between the variables can be demonstrated by a significant correlation between the two variables.
  • Scientific Research: To test hypotheses and examine the relationship between variables, correlation in statistics is frequently employed in scientific research. For instance, in the field of medicine, researchers may utilize correlation to investigate the connection between a certain drug’s effects on a patient’s health.

Check out our blog on Data Analytics tutorial to learn more about Data Analytics!

Data Science IITM Pravartak

Types of Correlation

Positive correlation, negative correlation, and zero correlation are the three primary types of correlation in statistics. Let us discuss each of them in detail:

Types of Correlation in Statistics

Positive correlation

When two variables go up or down simultaneously, there is a positive connection. In other words, as one variable’s value rises, the other variable’s value rises as well, and vice versa. Positive correlations have correlation coefficients that vary from 0 to 1, with a perfect positive correlation being equal to 1.

Negative correlation

When two variables move in opposing directions, there is a negative correlation. The second variable reduces as one variable rises, and vice versa. The correlation coefficient for a negative correlation spans from -1 to 0, with -1 denoting a perfectly negative correlation.

Zero Correlation

When there is no connection between two variables, there is a zero correlation. To put it another way, changes in one variable have no impact on changes in the other. When there is no association, the correlation coefficient is 0.

Correlation Coefficient Formula

Here are concise explanations of various correlation coefficient formulas:

Pearson Correlation Coefficient (r)

The Pearson correlation coefficient quantifies the linear relationship between two continuous variables. It is commonly used when both variables follow a normal distribution. The formula for calculating the Pearson correlation coefficient is as follows:

r = (Σ((X_i - X_mean) * (Y_i - Y_mean))) / (sqrt(Σ(X_i - X_mean)^2) * sqrt(Σ(Y_i - Y_mean)^2))


  • X_i and Y_i are the individual values of the two variables.
  • X_mean and Y_mean are the means (average values) of the two variables.

Spearman’s Rank Correlation Coefficient (ρ)

Spearman’s rank correlation coefficient is a non-parametric measure that assesses the monotonic relationship between two variables. It is suitable when the relationship is not necessarily linear but can be described by an increasing or decreasing pattern. It is calculated based on the ranks of the variables. The formula for calculating Spearman’s rank correlation coefficient is as follows:

ρ = 1 - ((6 * Σd^2) / (n * (n^2 - 1)))


  • Σd^2 represents the sum of the squared differences between the ranks of the paired data points.
  • n is the number of paired data points.

Kendall’s Rank Correlation Coefficient (τ)

Kendall’s rank correlation coefficient is another non-parametric measure used to determine the strength and direction of the relationship between two variables. Like Spearman’s coefficient, it employs ranks rather than actual values. Kendall’s coefficient is applicable for ordinal variables with a natural ranking or order.  The formula for calculating Kendall’s rank correlation coefficient is as follows:

τ = (n_c - n_d) / sqrt((n0 - n1) * (n0 - n2))


  • n_c represents the number of concordant pairs (pairs where the ranks have the same order for both variables).
  • n_d represents the number of discordant pairs (pairs where the ranks have opposite orders for the two variables).
  • n0 is the number of tied pairs on both variables.
  • n1 is the number of tied pairs only in the first variable.
  • n2 is the number of tied pairs only in the second variable.

Point-Biserial Correlation Coefficient (r_pb)

The point-biserial correlation coefficient examines the relationship between a continuous variable and a binary variable (dichotomous variable). It is employed when one variable is continuous (e.g., test scores) and the other is binary (e.g., pass/fail). The formula for calculating the point-biserial correlation coefficient is as follows:

r_pb = (M1 - M0) / sqrt((s1^2 + s0^2) / 2)


  • M1 is the mean of the continuous variable for the group with the binary variable value of 1.
  • M0 is the mean of the continuous variable for the group with the binary variable value of 0.
  • s1 is the standard deviation of the continuous variable for the group with the binary variable value of 1.
  • s0 is the standard deviation of the continuous variable for the group with the binary variable value of 0.

These correlation coefficients provide different perspectives on the relationship between variables and are chosen based on the nature and characteristics of the data being analyzed.

Get 100% Hike!

Master Most in Demand Skills Now !

Limitations of Correlation

A strong statistical technique that can help in understanding the link between several variables is a correlation. However, correlation in statistics has a number of drawbacks:

  • The existence of a correlation between two variables does not necessarily imply that one of them is the cause of the other; instead, correlation in statistics just shows that there is a relationship between them. The observed association can be caused by other causes.
  • Outliers are exceptional numbers that significantly deviate from the rest of the data, and they can affect correlation coefficients. Outliers have the potential to skew the correlation coefficient, leading to incorrect outcomes.
  • Correlation only assesses linear relationships between variables; non-linear relationships are not reflected. Correlation analysis might not be able to capture non-linear correlations if the relationship between variables is not linear.
  • Correlation coefficients are prone to variation depending on sample size. Less reliable correlation coefficients may result from smaller sample sizes.

Examples of Correlation

A crucial statistical tool for determining the link between two or more variables is a correlation. Here are some examples of statistical correlation:

Correlation Examples in Statistics
  • GDP and Unemployment Rate: The Gross Domestic Product (GDP) and unemployment rate are inversely correlated. The unemployment rate is typically low when the economy is doing well and the GDP is high, and vice versa. Economists frequently study this link to determine the economy’s overall health.
  • Blood Pressure and Body Weight: Body weight and blood pressure are positively correlated. A person’s blood pressure tends to rise along with their body weight. This link is frequently examined in medical research since it is crucial to understand the risk factors for cardiovascular disease.
  • Education and Voting Behavior: Voter turnout and educational attainment are positively correlated. People tend to vote more frequently as their education level rises. Political scientists have researched this link in great detail since it is crucial to understand vote trends.
  • Social Media Use and Mental Health: There is a link between using social media and poor mental health. A person’s chance of experiencing mental health problems appears to rise along with their use of social media. Recent years have seen research into this relationship as a result of worries about how social media affects mental health.

Prepare yourself for the industry by going through these Data Analyst Interview Questions now!


Correlation in statistics is a significant statistical tool that is useful in explaining the link between various variables for researchers. Researchers can create more precise predictions and draw more well-informed conclusions from data by researching correlations.

We tried our best to make you clear about the topic, but if you are still left with any doubts, reach out to our Community page!

Frequently Asked Questions (FAQs)

What is correlation with example?

Correlation measures the strength and direction of a relationship between two variables. For example, as study hours increase, grades tend to improve, indicating a positive correlation.

What is zero correlation?

Zero correlation is a type of correlation which indicates that there is no linear relationship between two variables. When one variable changes, there is no consistent pattern of change in the other variable.

What are the 4 types of correlation?

The four types are: Positive Correlation, Negative Correlation, Zero Correlation, and Partial Correlation, each indicating different relationships between variables.

What is correlation and regression in data science?

Correlation measures the relationship between variables, while regression predicts one variable based on another. Both are fundamental in data analysis and predictive modeling.

What is correlation and its types?

Correlation measures the relationship between two variables. Types include positive, negative, zero, and partial correlation, indicating different kinds of relationships.

What is a good example of a correlation?

A classic example is the positive correlation between height and weight; generally, taller individuals tend to weigh more.

What is the difference between correlation and regression?

Correlation measures the strength and direction of a relationship between variables, while regression provides a model to predict one variable from another.

What is correlation in Python?

In Python, correlation can be calculated using the corr() function from the Pandas library, which computes pairwise correlation of columns in a DataFrame.

How is correlation calculated?

Correlation is calculated using formulas like Pearson’s correlation coefficient, which divides the covariance of the variables by the product of their standard deviations.

Why is correlation important?

Correlation helps in understanding the relationship between variables, which is crucial for data analysis, hypothesis testing, and predictive modeling in various fields.

Course Schedule

Name Date Details
Data Scientist Course 01 Jun 2024(Sat-Sun) Weekend Batch
View Details
Data Scientist Course 08 Jun 2024(Sat-Sun) Weekend Batch
View Details
Data Scientist Course 15 Jun 2024(Sat-Sun) Weekend Batch
View Details

About the Author

Principal Data Scientist

Meet Akash, a Principal Data Scientist who worked as a Supply Chain professional with expertise in demand planning, inventory management, and network optimization. With a master’s degree from IIT Kanpur, his areas of interest include machine learning and operations research.