8.10 Target population

The target population in a survey is the specific set of people or other sample units about whom information is sought. For example, school authorities may wish to find out the food preferences of students who use the school cafeteria. In this case, the target population would be all the students who might eat at the cafeteria.

A survey can only give reliable information about the target population if the sample is drawn in an unbiased way from that population. This is done by a selection process involving randomisation. There are several variants of such a process, the simplest being a simple random sample. 

It is important to distinguish what is known as the sampling frame  from the target population. Ideally, these should be the same but they may not be, due to design features in the survey. The sampling frame is the actual set of units from which the sample could be drawn. 

The sampling frame differs from the target population when some members of the population have no chance of being included in the survey sample.

Inferences based on survey data should be recognized as being strictly about the units in the sampling frame and not necessarily about the target population, unless the two are identical.

 

Example

A market research organization, wishing to obtain information about the shopping habits of the adult population of Australia, decides to make a random selection of postcodes and then to dial every residential telephone number within the chosen postcode areas. 

This type of sampling is called cluster sampling. The target population is the set of adults who make decisions affecting shopping. The sampling frame, however, is restricted to people who have a listed landline telephone number. The sampling frame is further restricted to those people who are willing to participate in the survey.

Worked Examples

QUESTION 1

QUESTION 2

QUESTION 3