Prob Methods of sampling

Probability
1. more likely to be representative of the population from which they are drawn than non-probability samples.
  1. likelihood that a participant is included in the sample is known for each subject in the population.
  2. More commonly used
2. Simple random
  1. each member of the population has an equal and independent chance/probability of being selected into the sample
  2. Important principles
    1. members of the population are selected one at a time
      1. independent of each other and without replacement
  3. i.e lottery draw iplacing each possible number in a container and selecting one at a time without looking until the desired number is reached.
3. Systematic
  1. dividing the sampling frame into a number of intervals, randomly selecting a starting point and then selecting one element from each interval in a systematic way.
  2. arranged in such a way that bias is introduced into the sample selection process.
    1. i.e class list contains sub-groupings of individuals whereby some members of the sub-groups are never selected.
  3. may select every 10th person on the sampling list.
    1. sampling frame is a class list arranged in alphabetical order by surname
    2. decide to select every fourth name on the list
      1. a starting point is randomly selected between one and four and then every fourth element is selected
      2. Sample frame= divided into intervals of four elements
Cluster
1. whole population being divided into groups or clusters and then a certain number of clusters are randomly sampled using simple, systematic or stratified random sampling.
2. i.e geographic area, hospitals, schools
  1. required number of clusters is selected you then sample all units within the clusters.
    1. interested in obesity in primary school aged children in Victoria.
      1. create a sampling frame of all Victorian primary schools and then randomly select clusters or schools
        would then invite all children to participate in the study.
3. very economical method of sampling. ABS uses cluster sampling to collect data for the National Health Survey.
Stratified
1. involves dividing the population into groups (for example, age)
2. calculating the number of participants needed in each group or strata so that the numbers reflect their proportion in the population.
3. ensures that important sub-groups are identified proportionally prior to random sampling.
  1. i.e in many community surveys people from some ethnic groups are under-represented.
    1. Conduct a community survey and want Italian population in a certain area where 20% are. So ensure that 20% are included in the survey to be a accurate rep.
4. randomly selecting participants into each stratum.

Next up

Prob Methods of sampling

Description

Resource summary

Similar

	Created by shirley.ha about 11 years ago