Main methods of sampling
If you would prefer to view this interaction in a new web window, then please follow the link below:
A common student mistake...
Do not mix up random and convenience samples. Selecting the first ten people who enter a shop is not a random sample as every member of the population (potential customers) the firm is interested in does not have an equal chance of being selected - only those entering the shop stand a chance of being chosen. Indeed, those buying from this shop may have similar characteristics not shared by others in the sampling frame.
This mistake is often made by students in their internal assessments and extended essays when surveys are carried out. It is often written that:
'I chose the first ten people at random'.
This is not an accurate description of a random sample, but of a convenience sample.
How many respondents should be in a sample? This is the realm of statistics, and it will be touched on later. The answer, though, is likely to be 'fewer than you think'. Political public opinion polls seeking voting intentions of several million voters are based on samples of approximately 1,000. They are also based on careful stratified sampling methods. They make predictions with claimed accuracy of about +/-2%. Not a bad accuracy for such a small sample.
How questions are phrased is critical to the accuracy of the results. The use of leading questions will clearly affect the survey's reliability. Questionnaires can be designed to get a specific result, so the developers need to be careful.
Methods of asking questions
There are many ways to conduct research as has been discussed earlier. These include:
- Unstructured interviews
- Structured interviews
- Telephone survey
- Consumer panels
- Focus group
The validity of the result will be determined by how many people are asked, how they are selected and the nature of the questions. We looked at many of these methods in the previous section.
Matching sampling method to population
It is important to match the sampling method used to the nature of the population being considered. If, like a school, the population is all in one place then the options are different to a situation where the population is geographically scattered (e.g. an online store of some sort). The table below summarises some of these options:
|Nature of population||Sampling methods available||Sampling methods not available|