Confidence Interval for the Standard Deviation σ
distpic_z.gifThe χ2 distribution is not symmetric and there is a different distribution for each sample size. Thus there are two critical values for the confidence interval.

Left and right critical values
of chi squared χ2L and χ2R

For right, left or two tailed hypothesis tests, critical values of χ2 can be found in Table A-4. First compute (α ÷ 2) and 1.000 – (α ÷ 2). Look for the same area values in the column titles and look in the row for the number of degrees of freedom. df = n - 1
Because the χ2 distribution is asymmetric the right critical value is not equal to the left critical value as they are in the z and t distributions.
Confidence Interval
cie1pz.gif
Because the χ2 distribution is asymmetric the limits of the confidence interval are computed directly without computing the margin of error.