A residual plot for a linear regression shows residuals scattered randomly around zero with no pattern and approximately equal spread. What does this indicate about the linear model?
- AThe r^2 value must be close to 1.0 since there is no pattern in the residuals.
- BThe residuals confirm that there is no relationship between x and y.
- CThe linear model appears appropriate; conditions of linearity and equal variance seem satisfied.Correct
- DThe model is not a good fit because the residuals should follow a clear pattern.
Why C is right
A random, patternless scatter of residuals around zero with roughly constant spread is exactly what we hope to see in a residual plot. It indicates that the linear model is appropriate for the data (linearity condition satisfied) and that the variability of residuals is constant (equal variance condition satisfied).
Why the other 3 choices miss
- A
- Incorrect because r^2 measures the proportion of total variation explained by the model. Residuals can scatter randomly with both small and large r^2 values — a random residual plot is consistent with any r^2 as long as the model form is correct.
- B
- Incorrect because the absence of a pattern in residuals only tells us the linear model fits well. If there were no relationship, the regression line would be nearly flat and r^2 would be near 0, but residuals could still scatter randomly.
- D
- Incorrect because a pattern in the residuals would indicate a problem with the model (such as nonlinearity). A random scatter with no pattern is a sign that the model is appropriate.