Pearson Correlation Calculator

Pearson correlation measures the strength and direction of a linear relationship between two numeric variables. It ranges from -1 to 1. A value near 1 indicates that larger values of one variable tend to pair with larger values of the other along a roughly straight-line pattern. A value near -1 indicates that larger values of one tend to pair with smaller values of the other. A value near 0 indicates little linear association.

Correlation is unitless because it standardizes both variables. That makes it easy to compare associations across different measurement scales, but it also means the coefficient does not tell you the slope or magnitude of change in original units.

Pearson correlation is specifically about linear association. Two variables can have a strong nonlinear relationship and still show a low Pearson correlation. For example, a U-shaped pattern may balance positive and negative slopes, making the overall linear measure close to zero.

A scatter plot is essential. The same correlation coefficient can hide very different data shapes, clusters, outliers, or changing variance. The number is a summary; the plot shows whether the summary is appropriate.

A high correlation does not prove that one variable causes the other. Both may be driven by a third factor, the direction may be reversed, or the pattern may be coincidental. Time order, experimental control, plausible mechanism, and confounder handling are needed before making causal claims.

This warning is familiar because it is important. Correlation is often the beginning of an investigation, not the end. It can identify relationships worth modeling, monitoring, or testing more carefully.

Outliers can inflate or deflate Pearson correlation dramatically. A single extreme point may create an apparent relationship in otherwise scattered data. Conversely, a valid but unusual point can weaken a real association. Robust checks and visualization help decide whether a coefficient reflects the main pattern or one influential observation.

Range restriction also matters. If data only cover a narrow slice of possible values, correlation may appear weak even when the wider population has a strong relationship. The sampled range determines what the coefficient can reveal.