Probability Samples

Types of probability samples:

• Simple random sample - everyone in the population has equal likelihood of being selected
  • The most effective, but often hardest to execute
• Stratified random sample - we create strata based on some factor and take a random sample from each strata
  • Protects against bad sampling
  • Decreases variance - increases precision in subgroups
  • Decreases cost sometimes
  • We can choose a proportion of each group, or base the proportion on the proportion of the subgroup in the population
• Cluster sample - Group observations based on how they are collected, randomly choose several groups then random sample in each cluster
  • Easier and decreases cost compared to SRS
  • Cluster is also called primary sampling unit (psu)
  • We can also have secondary sampling units if we cluster again within the psu
• Systematic sample
• Multistage sample