Suppose a Shewhart control chart is applied to a sample statistic for a process in a state of statistical control (i.e. a random process) with known mean µ and standard deviation ?. Assume now that after the implementation of the control chart, the mean of the statistic has shifted down by 1 standard deviation. What is the expected number of samples observed before the shift is detected? What is the probability of detecting this shift by the second sample after the shift has occurred?