Your data
Paste a table straight from Excel or Google Sheets, or drop a CSV file. Nothing is uploaded anywhere — the whole calculation runs inside this browser tab.
CSV, TSV or plain text. Files saved in windows-1251 are detected and re-read automatically, so Cyrillic headers do not turn into garbage.
Parsing options
Columns
Result
Save and share
The drawing library is downloaded only when you press the button, and only on this page.
This page has no server side at all. The table you paste is parsed by JavaScript inside your own browser, the chart is drawn on a canvas element on your machine, and the export file is assembled locally. Nothing is uploaded, stored or written to any log. You can disconnect from the network after the page has loaded and everything will still work — which is the simplest way to verify the claim yourself.
Facts and limits of this method
| Range | −1 to +1; 0 means no linear relationship |
|---|---|
| What it measures | Linear association only |
| R² | The square of r — the share of variance in one column explained by the other |
| p-value | Exact, from the t-distribution with n−2 degrees of freedom |
| p-value method | Regularised incomplete beta function, not a normal approximation |
| Minimum rows | 3, and meaningfully around 20 |
When it misleads you
- Correlation is not causation, and this is not a formality. In the sample data ad spend and sales rise together, but both also rise through the year; a third variable driving both is the most common explanation of any strong correlation in business data.
- Pearson only sees straight lines. A perfect parabola — sales rising with price up to a point and falling after it — can produce an r near zero while the two variables are completely determined by each other. Always look at the scatter plot.
- A single outlier can create or destroy a correlation. One extreme point far from the cloud can drag r from 0.1 to 0.8, and removing it takes r straight back. Spearman correlation is the robust alternative.
- Statistical significance is not importance. With 10,000 rows an r of 0.03 has a tiny p-value and explains 0.09% of the variance — real, measurable and useless.
How it is calculated
The coefficient is the covariance of the two columns divided by the product of their standard deviations. That division is what removes the units and confines the result to the range from −1 to +1, which is why r can be compared across completely different pairs of variables.
Rows are used only when both values are present. A row with one number and one blank cell is dropped as a pair and counted, and the count of dropped rows is reported — silently keeping half a pair would bias the result in an invisible direction.
The p-value comes from converting r into a t statistic with n−2 degrees of freedom and evaluating the exact tail probability through the regularised incomplete beta function. Many calculators use a normal approximation here; on small samples it understates the p-value and makes weak relationships look significant.
R² is reported because it is the interpretable number. An r of 0.7 sounds like "most of it", but R² of 0.49 says the second column explains just under half of the variation in the first, which is a much more honest summary.
Questions and answers
What is a strong correlation?
As a rough guide: above 0.7 strong, 0.4 to 0.7 moderate, 0.2 to 0.4 weak. But context dominates — 0.3 is impressive in social science and disappointing in physics.
My p-value is significant but r is tiny. What does that mean?
That you have a lot of rows. Significance says the relationship is probably not zero; it says nothing about whether it is large enough to act on. Look at R² for that.
Pearson or Spearman?
Pearson when the relationship looks linear and there are no extreme outliers. Spearman when the relationship is monotonic but curved, or when outliers are present — it uses ranks and is much harder to distort.
Why did my correlation change so much after removing one row?
Because Pearson is highly sensitive to outliers. If one row can move r substantially, the correlation is a property of that row rather than of the dataset. Check the scatter plot and consider Spearman.
Is my data uploaded?
No. The coefficient and the p-value are both computed in your browser.