Paste your data
One variable per column, with a header row. Paste straight from Excel, Google Sheets, or a CSV - tabs, commas, and semicolons are all detected automatically.
What a correlation matrix tells you
A correlation matrix summarizes the pairwise linear (or rank) association between every variable in your dataset. Each coefficient runs from -1 (a perfect negative relationship) through 0 (no linear relationship) to +1 (a perfect positive relationship). It is usually the first table in the results section of a quantitative paper because it shows, at a glance, which variables move together before you fit any model.
Pearson vs Spearman
The Pearson coefficient measures straight-line association and assumes roughly continuous, normally distributed variables. The Spearman coefficient correlates the ranks instead of the raw values, so it captures any monotonic relationship and is robust to outliers and non-linearity. When a Pearson and Spearman estimate disagree sharply, that is a useful signal that the relationship is non-linear or outlier-driven.
Reporting correlations in APA style
APA style asks for a lower-triangular table with variable means and standard deviations, correlation coefficients reported to two decimals without a leading zero, and asterisks denoting significance. The APA export here follows exactly that convention. If you then want a colored correlation heatmap for a figure rather than a table, the heatmap chart page and the correlation heatmap use case show how to generate one with Python.
Frequently asked questions
How do I make a correlation matrix from my data?
Paste your data with one variable per column and a header row in the first line. The tool auto-detects whether you pasted tabs (from Excel/Sheets), commas, or semicolons, finds the numeric columns, and computes the full pairwise correlation matrix when you click Build.
Should I use Pearson or Spearman?
Use Pearson when the relationship between variables is roughly linear and the data are continuous and reasonably normal. Use Spearman (a rank correlation) when relationships are monotonic but not linear, when you have ordinal data, or when outliers would distort Pearson. You can switch between them and recompute instantly.
What do the stars and p-values mean?
Each off-diagonal cell shows the correlation coefficient with significance stars: * means p < .05, ** means p < .01, and *** means p < .001. Hover any cell to see the exact two-tailed p-value and the complete-case sample size for that pair. p-values come from the t approximation for a correlation coefficient.
How does the APA export work?
The APA copy produces a lower-triangular correlation table that also includes each variable's mean (M) and standard deviation (SD), exactly as journals expect. Paste it into Word and use Insert > Table > Convert Text to Table (tab-delimited). The LaTeX export gives you a booktabs table you can drop straight into a manuscript.
Is my data uploaded anywhere?
No. The entire calculation runs in your browser with JavaScript - your numbers never leave your computer and nothing is stored. That makes this safe for unpublished or sensitive research data.
Related tools
From a table to a figure
Plotivy turns the same data into a publication-ready correlation heatmap or coefficient plot - and hands you the Python code behind it.
Open the Analyze tool