I collect $n$ ($<20$) i.i.d. observations from any distribution. In order to compute the sample variance, I take $$s^2=\sum_i \frac{(\bar{X}-X_i)^2}{n-1}$$

If I want to build a confidence interval for the mean, I need the standard error of the sample mean. Do I divide by $n$, or by $n-1$ again?

$$S.E. \stackrel{?}{=}\sqrt{\frac{s^2}{n-1}}$$

  • $\begingroup$ The first part of @Macro's answer here does the necessary derivation, and I think it's covered by the answers here. $\endgroup$
    – Glen_b
    Jun 5, 2013 at 8:28

1 Answer 1


The sample mean is given by:

$$\bar{x}=\frac{\sum x_i}{n}$$

Thus, the standard error of the sample mean is:

$$SE = \sqrt{V(\bar{x})}$$


$$V(\bar{x})=V(\frac{\sum x_i}{n})=\frac{\sum V(x_i)}{n^2}=\frac{nS^2}{n^2}=\frac{S^2}{n}$$

Therefore, you divide by $n$ and not by $n-1$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.