Data Collection & Sampling

• Census: measures every member of the population
• Census advantage: accurate result
• Census disadvantage: expensive, may destroy units you're talking about
• Sampling units: individuals of the population
• Sampling frame: a list of sampling units

  e.g. a list of all students in a school
• Simple random sampling: everything has same chance of being selected
• You can generate a simple random sampling by either using a random number generator or lottery sampling
• Simple random sampling advantage: no bias
• Systematic sampling:

  • take every kth unit
  • $k=$$\frac{population}{sample}$ is how you calculate k
  • Pick a random number between 1 and k to be the starting point
• Systematic sampling disadvantage:

  Need a frame
• Stratified sampling:

  • the sample represents the groups (strata) of a population
  • To get group size do $\frac{sample}{population}\cdot strata$ for each strata and then pick randomly
• Stratified sampling advantage: reflects the population
• Stratified sampling disadvantage: population must have been classified by strata already
• Quota sampling: like stratified sampling but stratas are filled by interviewer / researcher instead of picked randomly
• Quota sampling advantage: no sampling frame required
• Opportunity sampling is when you take advantage of what's in front of you e.g. you're outside a supermarket and pick some people to try a product
• The quota in opportunity sampling is filled by those available at the time
• Opportunity sampling disadvantage: unlikely to be representative
• Qualitative data is non-numerical
• Quantitative data is numerical
• Quantitative data can be either discrete or continuous