Sampling

Question 1

Simple random sampling

Answer 1

• every member of the parent population is equally likely to be selected
• requires a sampling frame
• population must be sampled from 1-n
• random numbers generated
• if repeats occur, more random numbers must be generated

Question 2

Stratified sampling

Answer 2

• divide population into strata (different groups)
• all strata must be sampled
• selection of items to be sampled within each stratum is done at random
• usually accurate about the entire population + gives useful information about individual strata

Question 3

Proportional stratified sampling

Answer 3

• the numbers selected from each of the strata are proportional to their size

Question 4

Systematic sampling

Answer 4

• choosing individuals from a sampling frame
• choose from a random starting point and then every subsequent kth value from 1-n
• beware of cyclic patterns

Question 5

Quota sampling

Answer 5

• population divided into groups based on characteristic
• a quota of items/people in each group is set to try and reflect the group’s proportion in the whole population
• interviewer selects actual sampling units - non-random

Question 6

Opportunity sampling

Answer 6

• convenience sampling

- cheap method of choosing a sample where the sample is selected by simply choosing people who are readily available

Question 7

Cluster sampling

Answer 7

• starts with sub-groups of the population, but the items are chosen from one or several of the sub-groups
• sub-groups = clusters
• cluster should be reasonably representative of the entire population
• choose from a limited number of clusters

Question 8

Self-selected sampling

Answer 8

• people volunteer to be a part of the sample

- accepts any suitable volunteers