Unit 4 Flashcards
What is the major goal of research?
To gain information about different variables
When is a sample representative?
when it’s a good reflection of a wider population
-> results can be generalised
What are the factors that make a sample representative?
Suitability
Sample size Accessibility
Suitability
Inclusion and exclusion criteria based on demographic characteristics
Participant characteristics
major demographic characteristics
important top-specific characteristics (e.g. acheivement level in studies)
H0
Null Hypothesis
No difference between groups
H1
Alternative hypothesis
difference between groups
Why is sample size essential for the research design?
Increase in sample size -> increase in power
-> increases probability of rejecting null hypothesis
Are sample size and power linearly related?
No
-> too small ->low probability of rejecting null hypothesis
-> too large -> unnecessarily increases cost
=> after 90% slow increase in power for increasing sample size
Power
probability of correctly rejecting null hypothesis
doesnt statistically differ between study groups
What are accessibility sampling procedures useful for?
to describe:
- procedures for selecting participants
- setting and location of data collection
- agreements and payments to participants
- institutional review board agreements, ethical standards met, safety monitoring
Probability sampling
each population element has known chance of being chosen (above 0)
Non-probability samples
researcher can’t specify probabilities
What does probability sampling require?
a sampling frame
-> list of all members of population
Subtypes of probability sampling
Simple random sampling
Systematic sampling
Cluster sampling
Stratified sampling
Simple random sampling
Properties:
- population of N objects included in sampling frame
- sample of n objects
- all possible samples of n equally likely
Systematic sampling
list of population created
every kth member selected
-> e.g. every 3rd person
Cluster sampling
every member of a population assigned to one group (cluster)
one cluster used as sample (selected via other probability method)
only individuals within cluster surveyed
doesnt require sampling frame
stratified sampling
population divided into strata (groups)
within each group probability sample selected
What’s the difference between cluster sampling and stratified sampling?
Stratified sampling: includes elements fro. each stratum
Cluster sampling: only elements from sampled cluster