It would be too long and cost too much money to gather information from every single member in the target population when carrying out research so businesses collect it from a sample of people in the population. However, it is important the views of the sample are represenative of the whole. Sample size affects accuracy of the information: larger=more representative.
- Random sampling: respondents are selected for interview at random meaning all members have an equal chance of being chosen. However, it assumes all members are the same which is not the case so if sample is small it might not have the characteristics of the population.
- Quota sampling: target population is segmented into a number of groups that share specific characteristics and then a certain number of members are chosen from each group to be interviewed. It is more cost effective and useful where the proportions of the different groups within the population are known but results aren't statistically representative and population not chosen randomly.
- Stratified sampling: the sample is divided into segments based on how the population is divided up and then a random sample is chosen from each group but the proportions in the sample must the same as in the population as a whole.
Comments
No comments have yet been made