Add points to a scatter plot and watch the correlation coefficient and regression line respond as you do.
Your points
One pair per line as x, y. 5 points read. Click on the plot to add a point.
Scatter plot
r
0.9933
r²
0.9866
Slope
4.0789
Intercept
-30.3421
With 5 points, the correlation is r = 0.993, a very strong positive linear relationship. The line is ŷ = -30.342 + 4.079x, and it accounts for 98.7% of the variation in y. Try the Curved preset: y is completely determined by x, yet r is about zero, because r only measures straight-line association. Try the outlier preset to see one point wreck a strong correlation.
How to use this tool
Choose a preset, or type your own x, y pairs, one per line.
Click anywhere on the plot to add a point, and remove the last point with the button.
Watch r, r², the slope, and the intercept update after every change.
Try the outlier and curved presets to see what a correlation coefficient can and cannot tell you.
What does the correlation coefficient r tell me?+−
It measures the strength and direction of a straight-line relationship, from −1 (perfect negative) through 0 (none) to +1 (perfect positive). It does not measure curved relationships, and it does not show that one variable causes the other.
Why is r near zero for the curved preset even though y depends exactly on x?+−
Because r only detects linear association. In that data y = x², a perfect but U-shaped relationship. The rising right half and falling left half cancel out, leaving r near zero. Always look at the plot.
How can one point change r so much?+−
Points far from the others in x have high leverage and can pull the line toward themselves. In the outlier preset, one extra point turns a near-perfect correlation into a weak one.
What is the line drawn on the plot?+−
It is the least-squares regression line, the straight line that minimizes the sum of the squared vertical distances from the points to the line.