Residuals are nothing but errors. A residual is the difference between the observed value and the predicted value.

Residual = y - \widehat{y}

The given equation is

\widehat{y} = 8.5 + 69.5 x

Plugging in x = 2, gives us \widehat{y}=8.5+69.5\times 1=78

So when the number of phone lines are 2, the cost is $78.

The observed value is given as $150.

So the residual = $150 - $78 = $72.

**The residual for 2 phone lines is $72.**

The residual for 2 phone line is 72 dollars more than predicted.

Similarly, plugging in x = 5, in the least squares regression line gives us,

\widehat{y}=8.5+69.5\times 5 = 356

So, when the number of phone lines are 5, the cost is $356.

The observed value is $350.

So the residual is $350 - $356 = -$6.

**The residual for 5 phone lines is -$6**.

The residual for 5 phone lines is 6 dollars less than predicted.

I found an answer from www.stats4stem.org

**Least Squares Regression Line**

The **Least Squares Regression Line** is the line that minimizes the sum of the **residuals** squared. · The **residual** is the vertical distance between the observed point ...

For more information, see **Least Squares Regression Line**