What this site does. You bring a table of numbers, it gives you
numbers back. This page computes the Pearson correlation coefficient r for every
pair of your columns - one number from −1 to 1 for how closely two columns follow
a straight line - and shows it as a scatter plot and a heatmap. It runs entirely
in your browser — nothing is uploaded, nothing is stored. If you want the
reasoning rather than the answer, the correlation
article explains every number on a worked example.
r sees straight lines only: a perfect curve can score 0. Look at the scatter plot
before trusting the number, and never read r as a cause. Spearman, Kendall,
p-values and confidence intervals are on the
advanced calculator.
- Paste your data into the Input grid — straight from a
spreadsheet, or as CSV/TSV text — or import a CSV/TSV file.
Column x, then up to four y columns. Up to 5,000 rows. Every column counts as
a variable, x included.
- A cell that is not a number is skipped for its own column only. Each pair uses
the rows where both of its cells are numbers, and needs at least 3; a column
with one repeated value has no correlation.
- Press calculate. A change to the grid clears the result until the next press.
The status line names the strongest pair; see more -
hover or click - lists every pair.
- The upper chart shows every column against x, each scaled to
its own mean and spread, so columns of different sizes share one chart. It
follows the grid as you edit.
- The scatter plot shows one pair: pick a column on each axis,
and its r sits in the heading. 95% ellipse in the toolbar
draws the ellipse that holds 95% of a normal cloud with that pair's means and
covariance.
- The heatmap shows r for every pair: blue for positive, red for
negative, white near 0. Click a cell to show that pair in the
scatter plot.
- Read the result. Under the charts, pick a pair: r and N, the
rows it used. The Output grid lists every pair with N and r.
Export saves it as a CSV/TSV file.