5.5 Interpreting Y-Intercept in Linear Regression

What the y-intercept means

In a regression equation ŷ = a + bx, the y-intercept (a) is the predicted value of the response variable when the explanatory variable x = 0. It represents the point where the regression line crosses the y-axis. Interpretation should always be in context, using the actual variable names, not just 'x' and 'y'.

Context matters

A good interpretation follows a template: 'When [x variable] is 0, the predicted [y variable] is [a].' For example, if predicting weight from height, the intercept is the predicted weight when height is 0 inches — which is often not meaningful in real life, since x = 0 may fall outside the observed data.

Watch for extrapolation issues

Sometimes x = 0 is impossible or nonsensical (like zero height or zero age), making the y-intercept have no real-world interpretation — it's just a mathematical necessity for the line's equation. Always check whether 0 is a reasonable, observed value of x before interpreting the intercept meaningfully.

Key terms

Y-intercept (a) in ŷ = a + bx

The predicted value of y when x = 0.

How to interpret the y-intercept in context

'When [x variable] is 0, the predicted [y variable] is [a value].'

When might the y-intercept be meaningless?

When x = 0 is outside the range of the data or not realistic (e.g., zero age, zero height).

Role of the y-intercept in the regression line

It's the point where the regression line crosses the y-axis (x = 0).

Extrapolation and the y-intercept

Interpreting the intercept when x = 0 wasn't observed in the data is a form of extrapolation and may be unreliable.

Try an AP-style question

Which interprets the y-intercept?

1. The average score of the students in the data is 12
2. A student who studies 0 hours has a predicted score of 12
3. Every student in the data scored at least 12 points
4. Each additional hour of study adds 12 points