Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
t distribution and standard normal distribution
Message posted by Robin (via 209.56.125.161) on June 10, 2001 at 7:24 PM (ET)
What are the similarities and differences between the t distribution and the standard normal distribution. And if I wanted to construct a confidence interval, when would I use one rather than the other. I'm a little fuzzy on these two terms.....Thanks.
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: t distribution and standard normal distribution
Message posted by JG (via 128.8.22.137) on June 11, 2001 at 12:49 AM (ET)
Z=(X-mean)/(true SD) while t=(X-mean)/s where s=estimate of the true SD . The t distribution dependes on degrees of freedom which is how much data was used to estimate s . In the simplest case df=(n-1) . The t and Z distributions look the same except that t has a bigger variance. When df>30 they are very similar and when df>120 they are generally considered to be the same.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.