Sampling and Generalizability

These flashcards cover key concepts related to sampling and generalizability in research, including definitions, methodologies, and important distinctions.

What is sampling?

Selecting a subset of individuals from a population to estimate characteristics of the entire population.

What is external validity?

The extent to which study findings generalize to other people, settings, times, and places.

What determines external validity more: sample size or sampling method?

The sampling method; representativeness depends on how participants are selected, not how many.

When is a sample considered representative?

When all members have an equal chance of selection and key traits are proportionally reflected.

What is sampling bias?

Systematic error where some population members have higher selection probability than others, skewing results.

What is convenience sampling?

Recruiting those easiest to access (e.g., SONA students, clinic patients), which risks bias against harder-to-reach groups.

What is self-selection bias and how can titles worsen it?

a significant difference in characteristics between individuals who volunteer for a study and non-participants.; value-laden study titles attract skewed participants and reduce representativeness.

What is probability (random) sampling?

Any method giving every population member an equal, known chance of selection to maximize external validity.

What is simple random sampling?

Selecting individuals purely by chance (e.g., random number generator), giving everyone equal odds.

What is a drawback of simple random sampling?

Often impractical without a complete list; participation can still be voluntary.

What is systematic sampling?

a probability sampling method where researchers select members from a larger population at a fixed, periodic interval after choosing a random starting point.

• Selecting every nth person after a random start (e.g., start at #2, pick every 3rd).

What is cluster sampling?

Randomly selecting entire groups (clusters) and including everyone within selected clusters.

What is multistage sampling?

Randomly selecting clusters first, then randomly selecting individuals within those clusters.

What is stratified random sampling?

Dividing the population into meaningful strata (e.g., age, gender) and randomly sampling within each in proportion.

What is oversampling and why use it?

Intentionally sampling more from rare subgroups to get precise estimates; results are later weighted.

How do random sampling and random assignment differ?

Random sampling boosts external validity; random assignment boosts internal validity within experiments.

When might internal validity be prioritized over external validity?

In causal tests of mechanisms (e.g., drug effects), where random assignment is key.

When is nonrandom sampling acceptable?

When external validity isn’t the main goal (exploratory work, mechanism tests, or rare populations).

What is purposive sampling?

Nonrandomly recruiting people who meet specific criteria relevant to the study.

What is snowball sampling?

Using participant referrals to recruit others from rare or hidden populations; useful but network-biased.