Lecture 2 Flashcards
What are the two types of errors associated with surveys?
1 - Sampling errors
2 - Non-sampling errors
How can we control sampling errors?
Use of probability sampling methods and sample size
What can result in non-sampling errors?
None-response, inaccurate responses and selection bias
How can we control non-sampling errors?
Careful planning in the design stage
What are the two types of sampling procedures?
Probability sampling and non-probability sampling. In probability sampling some form of randomisation is used to generate the sample.
What are the four main types of probability sampling?
1 - Simple random sampling
2 - Stratified sampling
3 - Cluster sampling, and
4 - Systematic sampling
Describe simple random sampling.
Create a sampling frame, randomly select from this list. Works best if the sample frame is homogenous.
Describe stratified sampling.
This is used when we know there is likely to be systematic variation in the population of interest. It is a two stage process, firstly using the sample frame the population is divided into strata. Next a random sample is taken from each of these strata.
Describe cluster sampling.
Here, groups or clusters are identifies and then a random sample of these is selected.
What is the advantage of using stratified random sampling over simple random sampling?
It increases the representativeness of the sample.
What is multi-stage sampling?
“two-stage sampling process” i.e. cluster sampling first then systematic sampling.
What is convenience sampling?
Non-probability sampling, “man on the street”
What are the three most common examples of “purposive sampling”?
- quota
- snowball and,
- heterogeneous
What is quota sampling?
Involves the non-random selection of interviewees according to some fixed quota.
What is snowball sampling?
Sampling several people from the population of interest, and then asking to recommend others they know who might also meet the criteria for inclusion in the study.
