0

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.

Da Silva Lionel
  • 11
  • 4

1 Answers1

0

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.

Da Silva Lionel
  • 11
  • 4
  • Please search our site for [30](http://stats.stackexchange.com/search?q=30+answers%3A1) to read the many reasons why this advice needs to be taken only with careful thought and additional analysis. The highly voted thread at http://stats.stackexchange.com/questions/2541 looks particularly useful. For a realistic, important, and dramatic example of how this advice can fail, see http://stats.stackexchange.com/questions/69898 . – whuber Jul 06 '16 at 21:47