Sampling, bias and random numbers

We usually want information about a whole population but can measure only a sample. The selection method decides whether the sample can represent the population.

To explain why a method is unsatisfactory, identify the affected group and state how its chance of selection differs.

Method: interview the first 30 students entering the library.

Model answer: Students who do not enter the library at that time have no chance of selection, so the sample may not represent all students.

If 804 students are numbered 001 to 804, read random digits in groups of three. Reject 000, values above 804 and any repeated ID.