Probability Sampling

The essential characteristic of probability sampling is a procedure where each person or item is given a known chance of inclusion and the procedure is used for the selection of individual items in the Population. Possible methods are:

• Random sampling
  • Each individual has an equal chance of being selected.
  • Make sure more people are selected
• Stratification
  • If certain groups (or strata) can be identified before sample selection takes place it will be better to include this upfront. This is called stratification.
  • When you do not know the groups upfront you can identify those by asking a question:
    • "At the last general election which party did you vote for"
    • "Do you ride a motor cycle"
  • The retrospective use of information is known as post stratification.
• Cluster Sampling
  • If your population has a group or clusters which adequately represent the whole population it is more convenient and cost effective to select a random cluster and carry out a census for a complete cluster. This is called cluster sampling.
  • An interesting variant of cluster sampling is a random walk. For instance, conduct interviews with every 5th house.

With multistage design a series of related samplings are conducted.

Part of: M7-Quantitative Decision Methods-S1 - Managing Data and Survey Methods