Types of Correlations Explained: How to Identify Relationships in Your Data

Key takeaways

Positive and negative correlations are types of correlations that demonstrate a linear relationship between variables.

• The strength of a correlation appears as a correlation coefficient, which ranges from -1 to 1.

• Correlation is different from causation, because a relationship between two variables doesn’t automatically imply that a change in one variable causes a change in another.

• You can use statistical analysis techniques, like correlation analysis, in roles like data scientist and statistician.

Explore different types of correlations and real-world examples of each. Ready to start building skills like statistical analysis to extract insights from data? Earn an IBM Data Analyst Professional Certificate, during the course of which you have the opportunity to build skills in areas like model evaluation, data wrangling, programming, and more.

What are the different types of correlations?

A correlation describes the relationship between variables, whether it is positive or negative. If no linear relationship exists between variables, then the relationship has zero or no correlation. You express correlations as a figure, called a correlation coefficient, between -1 and 1, with a correlation closer to -1 or 1 than zero indicating a strong correlation. 

Positive correlation

When a positive correlation is present between variables, both variables will increase or decrease simultaneously. This means when one variable goes up, the other will also go up, and when one variable goes down, the other will also go down. For example, as temperatures rise in the summer, ice cream sales also rise, signaling a positive correlation between temperature and ice cream sales. Based on the strength of the positive correlation, the correlation coefficient can reach 1; any value above 0 and up to and including 1 indicates a positive correlation. 

Negative correlation

A negative correlation describes a relationship between variables where they move in opposite directions. As one variable increases, the other decreases, or as one decreases, the other increases. For example, when rainfall levels increase, sunglasses sales decrease. A strong negative correlation appears as -1, while a weaker negative correlation corresponds to a correlation coefficient closer to zero.

Zero correlation

Zero correlation means no linear relationship exists between variables, so as one variable changes, the other doesn’t move in a similar manner. For example, the more gardening you do, the better you are at doing math. You can see how these two variables would predictably have no relationship with one another. Instances of zero correlation have a correlation coefficient of 0.

Strong vs. weak correlation

A strong correlation has a correlation coefficient close to -1 or 1, while a weak correlation coefficient is close to 0. In a strong correlation, the variables will consistently move in the same or opposite direction. The less consistent the movement between variables, the closer the correlation coefficient is to 0.

How correlation is measured

You can measure correlation using a correlation analysis, which shows the direction and strength of the relationship and expresses them as a correlation coefficient. By using data analysis software like Microsoft Excel, you can simply input your data to perform a correlation analysis. Based on the correlation coefficient, you can interpret your results as follows:

• 1: Perfect positive correlation

• 0.7: Strong positive correlation

• 0.5: Moderate positive correlation 

• 0.3: Weak positive correlation

• 0: No correlation

• -0.3: Weak negative correlation

• -0.5: Moderate negative correlation

• -0.7: Strong negative correlation

• -1: Perfect negative correlation

The correlation coefficient can be any number between -1 and 1, but this description gives you a general idea of the potential strength ranges of your correlation.

Professions that work with types of correlations

Data analysts, data scientists, and statisticians rely on insights derived from statistical analysis to help inform critical decisions. Learn more about these roles and how they use statistical analysis techniques, such as correlation analysis, to perform research and make critical decisions:

• Data analysts: By implementing statistical tools and techniques alongside data mining, machine learning, and predictive analytics, data analysts can analyze data sets for insights to inform strategic business decisions.

• Data scientists: Similarly to data analysts, data scientists rely on statistical modeling and work with complex data using advanced programming skills, big data, and machine learning.

• Statisticians: By implementing statistical theories and models, designing surveys, and interpreting data, statisticians can identify key information from data sets to identify opportunities to improve processes.

What is an example of linear correlation?

Although you can find various real-world examples of linear correlation, in which two variables move in unison, it’s often challenging to find variables with correlation coefficients of -1 and 1, and instead, they usually fall somewhere in between. Explore examples of positive and negative correlation in real-world scenarios to better understand how they can appear.

Positive correlations:

• As your time spent working out increases, so does your water consumption.

• When the temperature increases, boat rentals increase.

• When your manufacturing costs rise, the production rate also increases.

• As doctors see more patients, more people receive diagnoses.

• As you spend more money on a marketing campaign, the number of new customers increases

Negative correlations:

• When the number of hours students spend studying for tests increases, the failure rate decreases.

• As poverty rates in a city increase, life expectancy decreases.

• When a business invests more money in onboarding, employee resignations decrease.

• As customer service complaints decrease, rates of returning customers increase.

What are the three main types of correlational analysis?

The three main types of  correlation analysis are Pearson correlation, Spearman correlation, and Kendall correlation. Pearson's correlation is the most common, and measures the linear relationship between variables. Spearman's correlation measures monotonic relationships that don’t necessarily move linearly. Kendall correlation is useful for nonparametric or categorical variables. 

Common mistakes when interpreting types of correlations

Correlation is frequently mistaken for causation, but it’s important to understand that a strong correlation does not automatically imply that one variable caused a change in another. Correlation means that you can find a relationship between the two variables and how they move, while causation means that one variable is responsible for that change. However, causation isn’t always present even when a strong correlation is, as sometimes an outside variable is causing the change. 

For example, a correlation analysis would likely show a positive correlation between snow tire purchases and snowboarding injuries. This doesn’t mean buying snow tires causes snowboarding injuries; instead, another variable is responsible for their increase, which, in this case, is the winter season.

Read more: Correlation vs. Causation: What’s the Difference? 

How to practice identifying types of correlations

To start practicing identifying types of correlations, you can use statistical analysis tools like Microsoft Excel and IBM SPSS Statistics [1,2]. To perform your correlation analysis, you will need data. To acquire usable data for your analysis, use sources such as Tableau or Data.gov, where you can access open government data and select the variables you’d like to measure in your analysis [3,4].

Explore our free data analysis resources

Join Career Chat on LinkedIn, where you can get updates on in-demand skills, popular tools, and relevant certifications. Or if you want to learn more about data analysis, check out these free resources:

• Watch on YouTube: 7 Essential Data Analytics Skills for Beginners

• Take the quiz: Which Data Analysis Course Should You Take? Find Out in 1 Minute

• Hear from an expert: 7 Questions with a Data Analytics Professor

Whether you want to develop a new skill or get comfortable with an in-demand technology, you can keep growing with a Coursera Plus subscription. You’ll get access to over 10,000 flexible courses. 

Build job-ready skills with Coursera Plus

Start 7-day free trial

• 
• 
• 
• 

Start 7-day free trial