The confidence interval test operates the same way as the BOUNDS test, but instead of considering the actual minimum and maximum of the reference data, it computes their standard deviation, and counts the values lying outside a confidence interval at a specified probability level, under the assumption that the data are a sample taken from a normal distribution. This level is calculated according to a Student distribution with the proper number of degrees of freedom. The default value of the probability level is 95%, which can be changed through parameters.