PSC 41 Validity And Sampling

External Validity

validity is the degree to which the evidence supports the conclusions or claims made about the data

Population

entire set of people or products that are of interest to a research

Sample

A subset of a population that is selected for inclusion in the experiment

Inference

Typically, conclusions are drawn back to the population

Non-representiative Sample

Selection of who is in the study is biased towards some characteristic (age, gender, education, cultural, background)

EX: Everyone in the population has an equal chance of being in the study

Representative/Probability

• Uses random sampling

  • Every member of the population has an equal likelihood of being chosen for the sample

Not Representative/Non-Probability

• Does not use random sampling

  • Different members of the population have different likelihoods being chosen for the sample

  • “Take what you can get” or target sampling

Probability Sampling

• All probability sampling strategies use random sampling

• All have some limitation built into how sampling occurs

Stratification Sampling

a probability sampling method that divides a population into distinct, non-overlapping subgroups (strata) based on shared characteristics like age, income, or gender, unbiased

Cluster sampling

usually biased that is a probability sampling method where researchers divide a large, geographically dispersed population into smaller, distinct groups called 

Convenience sampling

using a sample of people who are easy to contact and readily available to participate, biased

Snowball sampling

a non-probability, chain-referral technique where initial research participants recruit future subjects from their own social networks, biased

Systematic sampling

sampling is a probability sampling approach that selects sample members from a larger population at random but with a fixed, periodic interval, unbiased

Purposive sampling

sampling is a technique used in qualitative research to select a specific group of individuals or units for analysis, biased

quota sampling

a non-probability sampling method that relies on the non-random selection of a predetermined number or proportion of units

involves identifying subsets of the population and using nonrandom sampling strategies to fill a target number for each subset, biased