Skip to content
Grad Coach

Diagram · CC BY 4.0

Scatter plot examples: positive, negative and no correlation

Three panels of 30 simulated students: positive, negative and no correlation, each with Pearson’s r.

Three scatter plots of simulated data, each of 30 students, showing positive correlation, negative correlation and no correlation. All three share the vertical axis, exam score (%), from 20 to 100. Each panel is labeled with its Pearson correlation coefficient, r, computed from the plotted values.

Swipe sideways to see the whole diagram, or download the PNG.

Scatter plots showing positive, negative and no correlation. Derek Jansen, Grad Coach, CC BY 4.0.

What it shows

Three scatter plots of simulated data, each for 30 students, with exam score (%) on the vertical axis. In the first, exam score rises with hours studied per week; in the second, it falls with hours of paid work per week; in the third, it is unrelated to height.

Each panel is labeled with its Pearson correlation coefficient, r, computed from the plotted values. The data are illustrative, generated with a seeded random number generator, and each dot is one student.

How to read it

  1. Look at the direction of the cloud first. Rising from left to right is positive correlation (r = 0.89); falling is negative correlation (r = −0.79).

  2. In the third panel the points form no pattern, and r = 0.05, close to zero.

  3. The closer the points sit to a straight line, the stronger the correlation and the further r is from zero.

  4. Where two students share the same values, one dot is marked “×2”.

The numbers behind it

  • Positive panel: hours studied per week, spread across 0 to 20; score = 42 + 2 × hours + normal noise (SD 7); r = 0.89.
  • Negative panel: hours of paid work per week, spread across 0 to 20; score = 85 − 1.5 × hours + normal noise (SD 7); r = −0.79.
  • No-correlation panel: height normal with mean 170 cm and SD 9 cm; score drawn independently of it (mean 68, SD 10); r = 0.05.
  • Values are rounded to whole hours, centimeters and percent before r is computed; 30 students per panel.

Ways to use it in teaching

  • Cover the r values and have students rank the three panels by strength of relationship before revealing them.

  • r near zero rules out only a straight-line relationship; a curved one can also give r near zero, unlike the height panel, where there is no relationship at all.

  • Ask which of the two correlated pairs could plausibly be causal, and what a study would need to show it.

Correlation is the first step toward the tests in our guide to inferential statistics.

Sources

  • Çetinkaya-Rundel, M., & Hardin, J. (2024). Introduction to modern statistics (2nd ed.). OpenIntro. https://openintro-ims.netlify.app/
  • Pearson correlation coefficient. (n.d.). In Wikipedia. Retrieved October 18, 2026, from https://en.wikipedia.org/wiki/Pearson_correlation_coefficient
  • Scatter plot. (n.d.). In Wikipedia. Retrieved October 18, 2026, from https://en.wikipedia.org/wiki/Scatter_plot

Describing data

Related diagrams

All open educational resources

Need something else?

Tell us which diagram, template or worked examples would help your teaching, and we’ll do our best to add it to the collection.