Marketing : Sampling methods

Before an organization conducts primary research it has to be clear which respondents it wishes to interview. A company cannot possibly interview the whole population to get their opinions and views. This simply would be too costly and unfeasible. A sample of the population is taken to help them conduct this research. To select this sample there are again different methods of choosing your respondents, a mathematical approach called 'probability sampling' and a non- mathematical approach, simply called 'non-probability sampling'. Let’s look at these in a little more detail.

Probability Sampling Methods – A mathematical chance of selecting the respondent.

Simple Random Samples

With this method of sampling the potential people you want to interview are listed e.g. a group of 100 are listed and a group of 20 may be selected from this list at random. The selection may be done by computer.

Systematic samples

Out of the 100 people we talked about above, systematic sampling suggests that if we select the 5th person from the above list, then we would select every 5th, 10th, 15th, 20th etc. The pattern is the every consecutive 5th. If the 6th person was selected then it would be every consecutive 6th.

Multi-Stage Samples

With this sampling process the respondents are chosen through a process of defined stages. For example residents within Islington (London) may have been chosen for a survey through the following process:

Throughout the UK the south east may have been selected at random, ( stage 1), within the UK London is selected again at random (stage 2), Islington is selected as the borough (stage 3), then polling districts from Islington (stage 4) and then individuals from the electoral register (stage 5).

As demonstrated five stages were gone through before the final selection of respondents were selected from the electoral register.

Non Probability Samples

Convenience Sampling

Where the researcher questions anyone who is available. This method is quick and cheap. However we do not know how representative the sample is and how reliable the result.

Quota Sampling

Using this method the sample audience is made up of potential purchasers of your product. For example if you feel that your typical customers will be male between 18-23, female between 26-30, then some of the respondents you interview should be made up of this group, i.e. a quota is given.

Dimensional Sampling

An extension to quota sampling. The researcher takes into account several characteristics e.g. gender, age income, residence education and ensures there is at least one person in the study that represents that population. E.g. out of 10 people you may want to make sure that 2 people are within a certain gender, two a certain age group who have an income rate between £25000 and £30000, this will again ensure the accuracy of the sample frame again.

To summaries there are two types of sampling frames - probability and non-probability, and within these six types of sampling methods as discussed above.