Which variable sampling method is sensitive to variability and requires stratification of the population?

Mean per unit estimation (MPU) is indeed a method that is sensitive to variability and typically requires the population to be stratified for effective application. This method calculates the average value of the items in a sample and then uses that average to estimate the total value of the population.

When the population exhibits significant variability, the accuracy of the MPU method can be improved through stratification. By dividing the population into subgroups (strata) that are more homogeneous, auditors can ensure that each stratum is represented in the sample. This targeted sampling approach helps in obtaining more reliable estimates because it reduces the effect of variability by focusing on more uniform segments of the data.

Stratification allows for a better understanding of how different segments of the population behave, leading to more accurate estimates and, ultimately, a more effective audit process. This increased sensitivity to variability makes MPU particularly useful when the population has distinct groups with different characteristics or values.