Correlation calculator
Calculate the Pearson correlation coefficient between two paired data sets.
The correlation calculator takes paired X/Y values and computes Pearson's r, the standard measure of linear correlation, along with R² and a significance test (t-statistic and p-value) checking whether the correlation is meaningfully different from zero.
How to use
- Enter paired X and Y values, one pair per row — add as many rows as you have data points.
- Click "Calculate" to get Pearson's r, R², and a significance test (t-statistic and p-value) against r = 0.
Common use scenarios
- Checking whether two measured variables move together linearly.
- Verifying a correlation reported elsewhere against your own raw data.
- Assessing how strong a linear relationship is before building a regression model.
- Testing whether an observed correlation is statistically significant or could be due to chance.
✓Content & verification reviewed: 2026-08-25
How we verify Correlation calculator
How to confirm the correlation result matches what you'd expect.
r ranges from -1 to 1, with 0 meaning no linear relationship
A value near 1 means a strong positive linear relationship, near -1 a strong negative one, and near 0 means little to no linear relationship — though a low r doesn't rule out a strong non-linear relationship between the same variables.
The p-value tests against the null hypothesis that r = 0
A small p-value means the observed correlation would be unlikely if the true correlation were actually zero — it's evidence the linear relationship is real, not proof of how strong or how causal it is.
R² is simply r squared — the share of variance explained
R² tells you what fraction of the variance in one variable is linearly explainable by the other, ranging from 0 to 1 regardless of whether the correlation is positive or negative.
Before you rely on the result
- You've entered at least a handful of paired X/Y values, not just 2-3 points.
- You understand r measures only linear relationship — a non-linear pattern could still show a low r.
- You've checked the p-value alongside r rather than judging strength from r alone.
- You haven't concluded causation just because correlation is high and significant.
Who this is for
- Students and researchers checking a linear relationship between two variables.
- Anyone verifying a reported correlation against their own raw data.
- People assessing relationship strength before building a regression model.
- Analysts testing whether an observed correlation is statistically significant.
Real-world examples
Checking a relationship between two variables
A researcher enters paired measurements for two variables to see how strongly, and in what direction, they move together.
Verifying a reported statistic
Someone reads a correlation claim in a report and re-enters the underlying raw data to confirm the r value themselves.
Pre-checking data before regression
An analyst checks the correlation and its significance before deciding whether a linear regression model is worth building on the same variables.
Limitations
- The significance test assumes the underlying relationship is linear and the data is roughly normally distributed — a low p-value doesn't confirm causation, just that the linear correlation is unlikely to be zero by chance.
Frequently asked questions
What does a negative r value mean?
It means the two variables tend to move in opposite directions — as one increases, the other tends to decrease. The strength is still read from how close r is to -1.
Why is r low even though the data looks related on a chart?
Pearson's r only measures linear relationships — a strong curved or non-linear pattern can produce a low r even though the variables are clearly related.
What sample size do I need for a meaningful result?
There's no strict minimum, but very small samples (a handful of points) make both r and its p-value unstable and easily swung by a single outlier.
Does a significant p-value mean one variable causes the other?
No — a significant correlation only means the linear relationship is unlikely to be zero by chance. It says nothing about causation.
🔒 Runs in your browser — nothing is uploaded. Results are for reference only; see the disclaimer. Last updated: 2026-08-25
Subscribe for updates
New tools and tips, straight to your inbox. No spam, unsubscribe anytime.