# Confidence Intervals

Hi folks, Any advice on how to make things a little more understandable in this topic? Clues to look for in question or steps to solve problems that require to construct Confidence intervals? I am a bit unclear between the T-Distribution and the Z @/2 topic. Any advice would be much appreciated. Thanks!

use t when norm distribution population variance unknown non-norm distribution with pop variance unknown and large sample use Z when norm distribution pop variance know non-norm pop variance known and sample size is large large sample size= >30

… Memorize the intervals for significance level for both?

… Also what about small sample size, < 30 ?? Thanks!

For your first post after mine I usually just look it up in the back of the book in those charts. just remember to see if its a one-sided or two-sided with small sample sizes, if the distribution is normal then you just need to see if the pop variance is known is know. (If it is use z, if not t) for non-normal small samples, there is no appropiate test statistic

Thanks!

getterdone Wrote: -------------------------------------------------------> > for non-normal small samples, there is no > appropiate test statistic …unless the distribution is known.