MCRS: Validity & Sampling

External validity "something outside of our research"

Population validity To what extent does our research say something about what we are interested in?

Ecological validity Making general statements about a large group of people in different circumstances

Internal validity "Is the effect caused by the independent variable or something else? Concerns certainty with which you can test causal relationships with your research

Covariance Y must change as X changes

Temporal order (Time order) X must precede Y

Non-spuriousness Nothing except X must be influencing Y

Census Study of an entire population

High control High internal validity

Non-probability sampling Based on convenience for the researcher and provides insight and judgment

Probability sampling "Assigning the selection of sampling units to a mechanism over which the researcher has no control, so every unit has an equal chance of getting selected

Convenience sampling Non-probability, based on convenience for researcher

Purposive sampling (Non-probability)Specific units of analysis will meet a certain criteria of the researcher

Quota sampling (Non-probability) Attempts to replicate features in the sample that researcher thinks is important in the population

Network/Snowball sampling (Non-probability) Researchers relies on members of a network to introduce to other members of the networkk

Volunteer sampling (Non-probability) Participants recruited directly by researcher. "Excited" to participate

Sampling frame Master list from which a probability sample is selected

Random sampling (Probability) Selects units as "luck of the draw"

Stratified random sampling (Probability) Method to "force" members of a desired group into your sample

Systematic sampling (Probability) Sampling every nth person from a list

Multistage cluster sampling (Probability) Sampling in steps, funneling from state to individual household

Advantage Nonprobability Convenience

Disadvantage: Nonprobability Sample does not represent population being sampled

Advantage: Convenience sampling Speed, cost

Disadvantage: Convenience sampling Cannot generalize from sample to population

Advantage: Purposive sampling Meets a specific need of the researcher

Disadvantage: Purposive sampling & Quota sampling Sample may not represent the population

Advantage: Quota sampling Attempts to replicate features of the population in the sample. More readily done than random sampling

Advantage: Snowball sampling Identifies research participants the researcher might otherwise not be aware of

Disadvantage: Snowball sampling Depends on researcher's ability to network. Sample may over- or underrepresent aspects of population. Possible loss of diversity in sample

Advantage: Volunteer sampling Identifies willing participants researcher might otherwise not be aware of

Disadvantage: Volunteer sampling Agendas and interests of volunteers may influence research. Sample may over- or underrepresent characteristics of population

Advantage: Probability sampling Sample represents population being sampled

Disadvantage: Probability sampling Consumes time and resources

Advantage: Random sampling Ability to generalize from sample to population

Disadvantage: Random sampling May eliminate individuals who should be in the sample

Advantage: Stratified random sampling Minority groups that should be in the sample are forced into it

Disadvantage: Stratified random sampling Need to identify and sample a sampling frame for each subgroup to be represented in the sample

Advantage: Systematic sampling Only one or two random number starting points are needed to start sampling

Disadvantage: Systematic sampling Sample may over- or underrepresent features of the population

Advantage: Multistage Cluster sampling Eliminated need for a comprehensive sampling frame of every individual to be sampled

Disadvantage: Multistage Cluster sampling Sampling frame needed for each step of sampling. Sample may over- or underrepresent features of the population

Click

Type

Listen

Games

Drucken