Sampling Design Flashcards
What is a sampling unit (SU)?
Object of interest that is independent
What is the aim of sampling designs?
Provide best statistical estimates with the smallest possible confidence limits at the lowest costs
What are the types of sampling designs?
- Simple random sampling
- Stratified random sampling
- Systematic sampling
- Multistage sampling
What is simple random sampling?
Subset of statistical population where each member of the subset has an EQUAL probability of being chosen e.g. pick from hat, random number tables
What are the problems with simple random sampling?
Sample can end up not accurately reflecting the population it is representing (does not account for differences in population)
What is stratified random sampling?
Divides population into smaller groups based on heterogeneity area or different population densities
Does not require to sample randomly within each stratum can be done systematically
How to perform calculations for stratified random sampling?
- N = N1 + N2 + N3
N = size of area or population as sample units
Nh = size of stratum
h = # of possible sample units in stratum - Determine stratum weight (Wh) using Wh = Nh/N
- Decide on total number of samples to be taken, then multiply Wh for each stratum by # samples decided
What is systematic sampling?
Samples placed in a systematic fashion
Most common is centric systematic area-sample
What is centric systematic area-sample?
Most common systematic sampling
Where sample area is subdivided into equal squares and SU is taken from centre of each square
What are the advantages of systematic sampling?
- ease and simplicity of application
- desire to sample evenly across entire study area
- few samples available
What are the disadvantages of systematic sampling?
The existence of possible periodic trends in the study area although unlikely
What is multistage sampling?
It is taking a sub sample of sample
What are examples of biased sampling?
- accessibility sampling = restricted to samples readily accessible
- haphazard sampling = samples selected haphazardly (based on sampler)
- judgmental sampling = sampling selection based on prior experience
- volunteer sampling = samples obtained from volunteers
