standard deviation of sample mean = sigma/rt(n) --> called standard error!! or the sample mean Pr( -1.96 < Z < 1.96) = 0.95 Z= standard normal variable, in terms of sample mean, this is: Pr(-1.96 < (xgem - u)/(sigma/rt(n)) < 1.96) = 0.95 The (1-alfa) * 100% confidence interval limits for the population mean = xgem +/- Z(1-alfa/2) * (sigma/rt(n)) (2-sided interval) xgem +/- Z(1-alfa) * (sigma/rt(n)) (1-sided interval) We call this a confidence interval.