When should I use a regression plot?

Quantifying linear relationship between variables ● Prediction and forecasting ● Showing goodness of fit (R-squared)

When should I avoid a regression plot?

Non-numeric data ● When relationship is clearly non-linear and a line would mislead ● Categorical comparisons

What are common mistakes with regression plots?

Extrapolating far beyond the data range ● Assuming correlation implies causation ● Ignoring influential outliers that skew the fit ● Forcing a linear fit on data that is clearly curved (try a polynomial or LOESS instead)

Relationship

Regression Plot

Scatter plot with a fitted trend line showing the linear or polynomial relationship between two variables.

Study Hours vs Test Score

Linear fit with R² = 0.98

View data (12 rows)

Chart data table: Study Hours vs Test Score
Hours	Score
1	42
2	51
3	56
4	63
5	67
6	71
7	74
8	82
9	85
10	88
11	91
12	96

Make a regression plot with your data

Use a regression plot when…

Quantifying linear relationship between variables
Prediction and forecasting
Showing goodness of fit (R-squared)

Avoid when…

Non-numeric data
When relationship is clearly non-linear and a line would mislead
Categorical comparisons

Data it needs

Property	Value
Min Rows	8
Min Columns	2
Column Types	numbernumber

Visual anatomy

Marks

circleline

Channels

position-xposition-yline-slope

Axes

x-quantitativey-quantitative

Guiding principles

Consider instead

Scatter Plot

When the raw relationship matters more than a fitted line

Connected Scatter Plot

When data has a temporal ordering

Common mistakes

Extrapolating far beyond the data range
Assuming correlation implies causation
Ignoring influential outliers that skew the fit
Forcing a linear fit on data that is clearly curved (try a polynomial or LOESS instead)

History

Least-squares regression was developed independently by Legendre (1805) and Gauss (1809). Francis Galton coined 'regression' in the 1880s studying heredity.

Accessibility notes

Report R-squared, slope, and intercept as text and describe trend direction (positive/negative). Distinguish the trend line from the points by stroke weight or dash pattern, not color alone — color-vision-deficient readers may see the line and points as a single hue.

Got data? Let's see what works.

Drop your CSV. You'll get a Regression Plot plus four alternatives - ranked by which one actually fits your data best.

Try it free