← Math & statistics

Statistics Association

Correlation Calculator

Other languages English 日本語 Español 简体中文

Paste paired x,y data to calculate Pearson or Spearman correlation, review a readable scatter plot, and copy a concise summary for reports or review notes.

Use this page when your main question is the strength and direction of association. If you need fitted values, residuals, or model diagnostics, switch to linear regression after this first pass.

How to use

  1. Paste one x,y pair per row. Comma, semicolon, tab, or whitespace separators all work for simple two-column data.
  2. Choose Pearson for linear association or Spearman when you care more about monotonic ranking than straight-line fit.
  3. Read the coefficient together with the scatter plot and p-value, then decide whether you need regression or another follow-up tool.

Wave 2 statistics expansion

Pearson and Spearman from pasted XY data

Pearson measures linear association and pairs naturally with a straight-line fit.

Run a calculation to see the result summary.

Pearson r
Two-sided p-value
Sample count

    Scatter plot

    On narrow screens, ticks are thinned to preserve readability. Focus the chart and use ← / → to move between points.

    Run a calculation to inspect points.

    Correlation is not regression, and neither proves causation

    Correlation answers “how strongly do these variables move together?” Regression answers “what line best predicts y from x?” They are related, but not interchangeable. Start with correlation when the first decision is about association, then switch to regression if you need a fitted model.

    Pearson vs Spearman

    Pearson uses the original numeric values and focuses on linear association. Spearman converts each variable to ranks first, which makes it better when the relationship is monotonic but curved, or when a few outliers would otherwise dominate the result.

    Read the plot before trusting the coefficient

    A single coefficient can hide structure. Use the scatter plot to check for clusters, bends, outliers, or ceiling effects. If the pattern is clearly nonlinear, a high or low coefficient alone is not enough context for a decision.

    Frequently asked questions

    What is the difference between Pearson and Spearman correlation?

    Pearson measures linear association on the original values. Spearman converts values to ranks first, so it is better for monotonic but not perfectly linear relationships and is less sensitive to outliers.

    Does correlation mean causation?

    No. Correlation only describes how two variables move together in the observed data. A high coefficient does not prove that one variable causes the other.

    Should I use this page or linear regression?

    Use this page when your main question is the strength and direction of association. Use linear regression when you need slope, intercept, fitted values, residuals, or model diagnostics.

    Does the share URL include my pasted data?

    No. The share URL stores only lightweight settings such as Pearson or Spearman mode. The raw data stays in your browser.

    Comments (optional)

    To reduce load, comments are fetched only when needed.