What is a scatter plot?
A scatter plot draws one dot per row, placed by two numbers: one along the horizontal axis and one up the vertical axis. The pattern of the dots shows whether the two numbers are related, how strongly, and whether some rows break the pattern.
Typical questions: do students who score high in maths also score high in physics? Do stores with more staff sell more? Does more ad spend bring more sales?
How to read a scatter plot
- Direction: dots rising from left to right mean a positive relationship; falling means negative.
- Strength: the closer the dots sit to a straight line, the stronger the relationship.
- The trend line is the straight line that best fits the dots (least squares). Its slope tells you how much Y changes, on average, when X goes up by one.
- Outliers far from the cloud of dots can pull the line; look at them separately.
What the correlation coefficient r means
| r | What it suggests |
|---|---|
| 0.7 to 1 | Strong positive relationship |
| 0.3 to 0.7 | Moderate positive relationship |
| −0.3 to 0.3 | Little or no straight-line relationship |
| −0.7 to −0.3 | Moderate negative relationship |
| −1 to −0.7 | Strong negative relationship |
r only measures straight-line relationships. A curved pattern can have r near zero and still be a strong relationship, which is why you should always look at the plot, not just the number. And correlation doesn’t prove that one thing causes the other.
How to make a scatter plot in Excel
- Put the X values in the left column and the Y values in the right column.
- Select both columns and go to Insert → Charts → Insert Scatter (X, Y) or Bubble Chart → Scatter.
- Click + → Trendline to add a linear trend line.
- Choose Trendline → More Options and tick Display Equation on chart and Display R-squared value on chart. R² is r squared; take the square root, with the sign of the slope, to get r.
How to make a scatter plot in Google Sheets
- Select the two columns and choose Insert → Chart.
- Under Setup → Chart type, choose Scatter chart.
- Under Customise → Series, tick Trendline, set the label to Use equation and tick Show R².
Scatter plot or line graph?
Use a line graph when the horizontal axis is time and the order of the points matters. Use a scatter plot when both axes are measurements and each dot is an independent case, such as a student, a store or a day.