5.4 - Size and Types

A sample is a selection of a number of people whose views and opinions are considered to be representative to a much larger target group. Samples are required as it's not possible to collect information from every single person within your target group.

A chose sample group must be reliable, representative and accurate.

The sample size needs to be considered carefully as it is a small number representing the total population of a group. This is referred to as random sampling. In the course of conducting your own market research, a sample of people is acceptable, aslong as the members of the sample can be proven to be representative of a larger target group.

There are two methods used in sampling:

• Probability Sampling - Here it is assumed that in the representative population, every individual member has an equal chance of selection to be part of the sample group. Methods of this type of sampling include stratified sampling, random sampling, cluster sampling and systematic sampling. 
• Non-Probability Sampling - This is not based on a random selection of members of the sampling. Judgement sampling, convenience sampling, snowball sampling and quota sampling are examples of this type of sampling. 

Activity 19 - Why can't you simply choose people off the street to take part in a focus group?


There are various implications of using sampling methods:

With the use of different types of sampling, there are issues of reliability, cost and accuracy. Random sampling costs tend to be high because of it's large nature and scope. It can be time consuming, both to complete the survey and data analysis of the result.

There are sampling methods chosen as a result of their time efficiency. For such samples, accuracy is not paramount and hence the outcomes of such samples are not expected to be very technical and systematic. Whichever method of sampling is used, it should be a reflection of the group targeted for research and must be able to stand up to scrutiny.