Confidence Intervals are expressed for either a two-tailed test or a one-tailed test. A two-tailed test places half the confidence interval into the lower limit and half the confidence interval into the upper limit. A one-tailed test places the whole confidence interval into either the lower limit or the upper limit.

A 90% confidence interval for a two-tailed test would be 1.6449 standard deviations above and below the mean (top left picture).

A 95% confidence interval for a two-tailed test would be 1.96 standard deviations above and below the mean (top middle picture).

A 99% confidence interval for a two-tailed test would be 2.5758 standard deviations above and below the mean (top right picture).

A 90% confidence interval for a one-tailed test would be 1.2816 standard deviations above or below the mean (bottom left picture shows above the mean for upper limit test).

A 95% confidence interval for a one-tailed test would be 1.6449 standard deviations above or below the mean (bottom middle picture shows above the mean for upper limit test).

A 99% confidence interval for a one-tailed test would be 2.3263 standard deviations above or below the mean (bottom right picture shows above the mean for upper limit test).