0
$\begingroup$

I'm currently calculating some confidence interval for some mean value by group. Some of them has less then 30 individual inside.

Does it still make sense to calculate the confidence interval for those group using the student law ?

I know that before calculating the confidence interval for a proportion using the normal law you must assert that there is more than 30 individual in the sample but i really dont know about a mean and the student law.

Thanks in advanced.

$\endgroup$
5
  • $\begingroup$ Student law? As in student t distribution? $\endgroup$
    – Repmat
    Jul 6, 2016 at 19:19
  • $\begingroup$ Yes @Repmap. I think it is how it's called in english $\endgroup$ Jul 6, 2016 at 19:41
  • $\begingroup$ demarcheiso17025.com/images/stat16.gif I'm talking about the t distribution behind this table $\endgroup$ Jul 6, 2016 at 19:44
  • $\begingroup$ You should take a look at the assumptions section in the t-test wiki page. $\endgroup$
    – Mur1lo
    Jul 6, 2016 at 19:48
  • $\begingroup$ Thanks. But this is the prerequisite for a mean comparison. to test if a categorical variable has a strong bond with a numerical one. i dont know if it also apply if it is just for a confidence interval $\endgroup$ Jul 6, 2016 at 19:58

1 Answer 1

0
$\begingroup$

http://www.stat.wmich.edu/s216/book/node79.html#tci

I got the answers here:

"We end this section with a reminder that the confidence interval ( 7.5) works well ONLY when at least one of the following conditions hold: (i) the population histogram looks like the normal curve, or (ii) the sample size is relatively large ($n\geq 30$ is a frequently used thumb rule . The following table summarizes the performance of the t-intervals under four situations." I thanks everybody who tried to helps.

Best Regard.

$\endgroup$
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.