Correlation Matrix
Compute a Pearson correlation matrix across numeric CSV columns.
About this tool
Drop a CSV and get the Pearson correlation coefficient for every pair of numeric columns. Values near 1 or -1 mean a strong linear relationship, and values near 0 mean none. Non-numeric columns are skipped.
How to use
How it works
Pearson's r measures linear association: r = covariance(x, y) / (sd(x) × sd(y)). The matrix is symmetric with 1 on the diagonal. Correlation does not imply causation, and a single outlier can move r a lot.
r = Σ(xᵢ − x̄)(yᵢ − ȳ) / √[Σ(xᵢ − x̄)² · Σ(yᵢ − ȳ)²]
Worked example
Two columns with an outlier
- Input
- x: 1,2,3,4,5,6,7 y: 2.1, 3.9, 6.2, 7.8, 10.1, 12.2, 30
- Output
- x vs y: 0.861074
- Walkthrough
- y rises with x, so r is strongly positive. The last value 30 breaks the otherwise straight line and keeps r from being closer to 1.
Limitations
- Pearson only; no Spearman or Kendall rank correlation.
- No heatmap picture, only the numbers.
- Detects linear relationships only. A perfect curve can still give r near 0.
Questions
How do I make a correlation matrix from a CSV?
What is a good correlation value?
Does it handle text columns?
Is correlation the same as causation?
Features
- Pearson r
- All numeric pairs
- Runs in your browser
Related resources
Guides and videos from around the web—opens on YouTube or the publisher’s site.
SQL Tutorial - Full Database Course
freeCodeCamp.org
SQL and tabular data sit next to CSV/JSON data tools.
Watch on YouTubeMachine Learning for Everybody
freeCodeCamp.org
Data prep context before ML-oriented helpers.
Watch on YouTube
- Getting started with JSON Schema
JSON Schema
Validate structured data after converting formats.
- RFC 4180 — CSV
IETF
Common CSV conventions for parse and convert tools.