# Standard Error of Estimate

#### StatGeek

##### New Member
Hi,
In a simple regression analysis based on the least squares regression equation, I got the predicted value of 7.15 for the predictor value of 11. The sample size is 35. I computed the Standard Error of estimate (Sx|y), which was 1.17.

Now my question is that what this value tells us? Does it mean that 68% of the times the predicted value will be 7.15 ± 1.17? Or what?
My friend said 68% would be within one SEE, but how is it possible? In the given example, 7.15 is not the mean of the predicted variables (Y), nor is 1.17 standard deviation!

#### KaterinaM

##### New Member
Maybe this video can partly help you

Hi,
In a simple regression analysis based on the least squares regression equation, I got the predicted value of 7.15 for the predictor value of 11. The sample size is 35. I computed the Standard Error of estimate (Sx|y), which was 1.17.

Now my question is that what this value tells us? Does it mean that 68% of the times the predicted value will be 7.15 ± 1.17? Or what?
My friend said 68% would be within one SEE, but how is it possible? In the given example, 7.15 is not the mean of the predicted variables (Y), nor is 1.17 standard deviation!

#### hlsmith

##### Less is more. Stay pure. Stay poor.
I think of the SE as the estimated population standard deviation. So 68% of data are w/in 1 and 95% w/in plus or minus 2, etc.