W2 - Sampling Flashcards
Where do you draw your sample from?
Population of interest
What does the type of sampling method used depend on?
Research Q
Money + time available to spend on sampling procedure
Why sample?
Rarely have access to the whole population that were interested in
Define a sample
Group of participants selected from a larger population
Define inference
Extent to which the sample characteristics reflect those of the population
What are the types of sampling
Probability
Non-probability
Probability sampling
Selecting from a pop where each member has a chance of being selected.
Probability of selection is known.
Non-probability sampling
Researcher doesn’t know what population size is + doesn’t have access to it.
Probability of selection is UNKOWN.
What are the types of probability sampling
Simple random
Stratified random
Systematic
Cluster
What are the types of non-probability sampling?
Convenience
Snowball
Simple random sampling
All members of pop has equal chance for being selected
Problem with simple random sampling
Could be unrepresentative of the population
How would a sample being “unrepresentative of the population” be minimised?
Larger sample groups
Strengths to simple random sampling
Easy
Large samples likely to be representative
No info needed prior to sampling
Weakness to simple random sampling
Larger samples often needed = small samples could be unrepresentative
Comparisons between small subgroups may be difficult
Stratified random sampling
Stratifying pop into groups based on known characteristics.
Sample then drawn from each group
Why is a sample drawn from each group made in stratified random sampling?
To mirror proportional representation of the pop
What are the types of stratified random sampling?
Proportionate
Disproportionate
Proportionate stratified random sampling
Each sample size is directly proportional to the pop size of the entire pop
Disproportionate stratified random sampling
Each sample size is not proportional to the pop size of the entire pop
Strengths of stratified random sampling
Fewer people need to be sampled to achieve same representativeness
Can customise sample to be comparable to a wider population
Can make inferences about small subgroups
Weaknesses of stratified random sampling
Req obtaining info for the particular stratification variables
Becomes difficult if there’s several variables
When is systematic sampling typically used?
Very large pop
How does systematic sampling work?
Experimenter selects from an ordered list every 5th, 10th… entry starting from a random position
Cluster sampling
What is the sampling unit?
Cluster of ppl belonging to a naturally occurring group
i.e geographical
Non-probability sampling
Convenience sampling
Sample selected due to being accessible to the researcher.
Considered easiest, cheapest + least time consuming.
Non-probability sampling
Snowball sampling
When is it usually done?
Very small population
Hidden populations, diff for researchers to access
Non-probability sampling
Snowball sampling
How does it work?
Researcher asks initial participant to ID another pot participant who meets criteria
2) Which of the following research design would give us the most confidence in determining causality?
a. Cross-sectional design
b. One group pre- and post-test design
c. Randomised control trial
d. Longitudinal design
Randomised control trial
A cross sectional study is an example of what type of research design?
Observational
Which type of experimental design provides the most confidence in determining causality?
True experimental
