Systematic sampling is a method of selecting a sample by taking every k-th item from an ordered list, starting from a randomly chosen point within the first k items. The interval k is the size of the population, listed in a sampling frame, divided by the sample size. To pick 50 invoices from 2,000, an auditor sets k at 40, draws a random start between 1 and 40, say 17, and then selects items 17, 57, 97 and so on to the end of the list.
The method is quick, easy to document and spreads the sample evenly across the population, which makes it popular in audit work and surveys, where the gap is the audit sampling interval. Its main weakness is periodicity. If the list has a pattern that repeats at the same interval, such as weekly sales recorded in groups of seven, the sample can be badly biased. Random sampling is then safer.
Ref 1010001599