DATA - stays on your device
AI Assistant

Correlation Heatmap

See every pairwise correlation at a glance — red means related, blue means inversely related

A correlation heatmap is a matrix where each cell shows the Pearson r between two numeric columns. The color makes patterns pop instantly: red cells (r near +1) mean the two variables rise together; blue cells (r near -1) mean one rises as the other falls; white/pale cells (r near 0) mean no linear relationship.

This tool loads a sample dataset of 30 people with Age, Height, Weight, Score, and Income. Height and Weight are intentionally correlated (~0.7); Age and Income are moderately correlated (~0.6); Score is independent of the rest. Click Link Data to build the heatmap and see which pairs are actually related.

Paste any multi-column numeric CSV to build a correlation heatmap for your own data.

Graph

Popular Lessons

Word Problem Solver Visual Word Problem Solver Function Root Finder

FAQ

What is a correlation heatmap?
A correlation heatmap is a grid where rows and columns represent variables, and each cell shows the Pearson correlation coefficient r between that pair. Colors encode the strength and direction: warm colors (red/orange) for positive correlation, cool colors (blue) for negative, and neutral (white) for near zero.
What is Pearson r and what do its values mean?
Pearson r ranges from -1 to +1. r = +1 is a perfect positive linear relationship (as X increases, Y always increases proportionally). r = -1 is a perfect inverse relationship. r = 0 means no linear relationship — though a nonlinear relationship may still exist. As a rough guide: |r| > 0.7 = strong, 0.4–0.7 = moderate, < 0.4 = weak.
What does r = 0 mean — does it mean no relationship?
r = 0 means no linear relationship. Two variables can have a strong nonlinear relationship (e.g. U-shaped) and still produce r ≈ 0. Always pair the heatmap with scatter plots to check for curves or clusters that Pearson r misses.
When should I use a correlation heatmap?
Heatmaps are most useful when you have 3 or more numeric columns and want to quickly find which pairs are worth investigating further. They are a standard first step in exploratory data analysis (EDA) before building regression models — they help you spot multicollinearity and interesting predictors.
Does correlation mean causation?
No. A high correlation between two variables only means they move together in this dataset — not that one causes the other. Ice cream sales and sunburn rates are highly correlated (both peak in summer), but neither causes the other. Establishing causation requires controlled experiments or careful causal inference methods.
Why is the diagonal always 1.0?
The diagonal of a correlation matrix shows the correlation of each variable with itself, which is always a perfect +1. This is expected and normal — it just confirms the matrix is correct.

Related Topics

Regression CalculatorScatter Plot MakerStatistics Calculator