Week 6 Flashcards
What is the main idea behind sampling?
Representativeness
Gap between what is stated in the theory and what occurs in reality
Ideal-reality gap
The process of selecting a subset of cases in order to draw conclusions about the entire set
Sampling
The population to which researchers would like to generalize their results
Target population/define study population/population of interest
The entity about whom or which the researcher gathers information. What are the different types?
Unit of analysis
individuals, groups, social artifacts
For the following descriptions, name the unit of analysis:
A - Anything that is not a human being
B - Cancer patients
C - Women with breast cancer
A - social artifact
B - group
C - individual
Why do we want the unit of analysis to be clear?
We do not want to make false generalizations from one unit of analysis to another
Operational definition of the population that provides the basis for drawing a sample; i.e. a list of cases from which a sample may be selected
Sampling frame/accessible population
More concrete than the target population. Turns a target population into a tangible list
Sampling frame
What are the two reasons that we do not study every single person in a population?
Time and money
What is the only situation in which a researcher will take into account every single person?
Census data
A mismatch between the target population and sampling frame
Coverage error
What are the types of coverage errors?
Undercoverage - omission
Overcoverage
- duplication
- Wrongful inclusoin
What are the causes of coverage error?
Incomprehensiveness
Technical problem
Attrition
Which coverage error is more common?
Undercoverage > overcoverage
Quitting the study prematurely; more prevalent in longitudinal studies
Attrition
The deviation of the selected sample from the true characteristics, traits, behaviours, qualities or figures of the entire population
Sampling error
Characteristics of the population
Parameters
Estimates of population parameters
Statistics
Difference between sample statistics and population parameters
Sampling error
Sampling error is only applicable to _______ research
quantitative
If the statistic avg is 3.75, and the parameter avg is 3.72, what is the sampling error?
0.03
By reducing our sampling error, we increase the ______ of the study
validity
Information from the sample is linked to the population via the ______ _______
sampling distribution
Sampling distribution:
- __________ concept
- _______–> _______–> _________
Aka as the _____ _____
theoretical
sample –> sampling distribution –> population
Aka the bell curve
Helps us to understand the relationship between sample statistics and population parameters
Sampling distribution
A distribution of a statistic, such as a mean, across all random samples that could be drawn from a population
Sampling distribution
How do we obtain representativeness in quantitative research?
Random selection/sampling
Random selection/sampling:
- _____ selection bias
- _____ chance of being selected
no
equal
What is another word for probability sampling?
Random sampling
Everyone in the target population has an equal chance of being selected for a study
Random sampling
Describe the logic behind random selection
Random selection –> representativeness –> generalizability/external validity
What is the equivalent of generalizability in qualitative sampling?
Transferability
Qualitative sampling:
- Sampling for ________ ______, not for statistical reasons
- Trying to make sense of your data; trying to interpret the ______ given by the study participants
- _______ do not work
- Sampling for meaning = _________ sampling
- Once we hit ______ _______, we stop sampling
meaningful patterns meaning numbers theoretical theoretical saturation
The point at which no new themes emerge from the data and sampling is considered complete
Theoretical saturation
Study findings fitting outside that particular study
The possibility that findings would have meaning to another group or could be applied in another context
(qualitative)
transferability
Transferability:
- ________ generalization, NOT ________ generalization
analytical
statistical
Match the sampling logic to the research type:
A - to produce a representative group of participants so results from your sample can be generalized to the population
B - Less interested in a representative sample, but still need to find people to participate which fit the overall picture
A - quantitative
B - qualitative
What are the different types of sampling strategies?
Probability sampling
- simple random
- systematic
- stratified random
- multi-stage cluster
Non-probability sampling
- Convenience
- Quota/theoretical
- Purposive
- Snowball/network
Random selection where each element of the population has an equal and independent chance of being included in the sample
Random sampling
Provides the greatest confidence that the sample is representative, and unbiased
Probability sampling or random sampling
Why is random sampling rarely used?
Seldom have complete list of target population
Less generalizable because less representative samples, but more feasible for the researcher to obtain
non-probability sampling
How can we increase our confidence in our non-randomly selected sample to be representative of the population?
Carefully use inclusion/exclusion criteria
A probability sampling procedure in which a population is divided into strata and independent random samples are drawn from each stratum
Stratified random sampling
When is it a good idea to use stratified random sampling?
When the target population is very diverse
Compare simple random and stratified random in terms of sampling error.
Simple random - subject to greater sampling error, especially with small sample sizes
Stratified random - reduced degree of sampling error; researchers can systematically control relevant sources of variability in the population
A probability sampling procedure in which the population is broken down into smaller areas called clusters, and a random sample of clusters is drawn
Multi-stage cluster sampling
The concept of hierarchy applies well to this probability sampling method
Multi-stage cluster sampling
Multi-stage cluster sampling is ______ efficient
cost
Non-probability sampling:
- Cases in the population are not ______ selected and they do not have an ____ ____ of being selected
- Results are not ______
- mainly used in ________ research
randomly
equal chance
generalizable
qualitative
A non-random sample in which the researcher selects anyone he or she happens to come across
Convenience sampling
A non-probability sampling method in which units are selected into the sample on the basis of pre-specified characteristics, so that the total sample will have the same distribution of characteristics assumed to exist in the population being studied
Quota sampling/theoretical sampling
A non-probability sampling method in which the researcher uses a wide range of methods to locate all possible cases of a highly specific and difficult to reach population
Purposive sampling
A non-probability sampling method in which the initial study participants may be asked to suggest additional people for the study
Snowball sampling
Will involve finding a gatekeeper who can introduce the researcher to other persons in a hard to reach population
Snowball sampling
What are the three major characteristics that exist in probability sampling, but not in non-probability sampling?
Random selection
Representativeness
Generalizability
A large sample without random sampling or with a poor sampling frame is ______ representative than one with random sampling and an excellent sampling frame
less
If we had to choose between sample size and a good sampling frame (i.e. representative), which would we prefer?
Representative > sample size
In which case would a sample size be very small?
Qualitative sampling - especially when finding a signature person
What are the major determinants of sample size in quantitative research?
The degree of accuracy required
The degree of variability or diversity in the population
The number of different variables examined simultaneously in the data analysis
What are the major determinants of sample size in qualitative research?
Research tradition used
Data saturation
Sampling methods depend on the ______ of research question as well as the _____ of your research topic - i.e. there is no _______
type
nature
What are key points to consider for recruitment of study participants?
Must be ethical, voluntary, incentives should not be too large, and the risks should be shared
