SAMPLING THEORY

1
SAMPLING THEORY

• 1. Random Sample
• Advantages
• All have an equal chance of being chosen.
• This avoids subjective bias and human element.
• Disadvantages
• Practical use restricted to small populations.
• Small samples may not be representative of the
  population. Therefore you need a large sample.
• Costly and time consuming.
• 2. Systematic/ Quasi-random sample
• Less random than 1. E.g. choose every 10th
  person. Widely used in the UK as it offers a far
  more practical approximation to random sampling.
• A number of methods are used e.g. polling
  records, telephone books, trade association lists
  etc.

2

• 3. Stratified random-sampling
• This sample is divided up into segments or
  strata, often preferred as it makes the sample
  more representative of the whole group.
• For example if a business wished know how class
  affects demand for food. Put the population into
  classes then take a random sample from these.
• 4. Quota sampling
• Population is segmented and interviewers are
  given targets/quotas.
• Advantages
• Cheaper
• Faster
• Administrative efficiency
• Useful when population has definite segments.
• Disadvantages
• Sample is not randomly chosen.
• Selection bias.

3

• 5. Cluster sampling
• Put population into clusters (often via
  geographic areas then take a random sample from
  the cluster).
• Advantages
• Useful if population is widely spread
• Cheap and fast.
• Disadvantages
• May miss sections of the population if only have
  a few clusters.
• 6. Multi-stage sampling
• Random systematic method
• Choose country at random
• City
• District
• Street
• Household
• This method is bad when it is widely dispersed
  and is also very expensive
• 7. Snowballing
• Highly specialised start with one individual or
  group and use these to contact others. Only used
  when sampling methods arent possible.