Chapter 7 - Sampling and Generalizability Flashcards
what does external validity consider?
sample and setting
define population
a larger group from which a sample is drawn, the group to which a study conclusion are intended to be applied
define census
a set of observations that contains all members of that population of interest
what is the difference in a biased vs. an unbiased sample
biased: sample in which some members of the population of interest are systemically left out and results cannot generalize to entire population of interest
unbiased: a sample in which all members are equally likely to be included (random) and results can be generalized to whole population
what are the ways that a sample can become biased??
- sampling those who are easy to contact (convenience sampling = easiest to acces/readily available)
- sampling only those who volunteer (self-selection = only having people that volunteered)
what is the difference between probability and nonprobability sampling?
probability/random: name for random sampling, drawn from population were each person has equal chance of being included in sample (simple random sampling, systematic, cluster & multistage, stratified, oversampling. external validity is priority
nonprobability/nonrandom: nonrandom sampling such as convenience, purposive, snowball, quota, that give a biased sample, causal
define simple random sampling
sample is chosen at random from popluation of interest, number generator
define systematic sampling
probability sampling tecnhqiue in which the researchers uses a randomly chosen number and count off every the member.
ex. chose 15, so they pick every 15th person
define cluster sampling
cluster of particiaptns whitin population are selected at random, followed by data collection from all individuals
selecting 10 highschool from all highschool in ontario
define multistage sampling
involves at least 2 stages, a random sample of clusters followed by a random sample of people within the selected clusters
seclting 10 people from each class out of the 10 schools you chose in your cluster sample
define stratified random sampling
random sampling where the researcher identifies particular demographic categories and then randomly selects individuals
to retain important charactieris of population in sample, example male female ratie
define oversampling
a variation of stratified in which the researcher intentionally overrepresents one or more groups. if one groups has very low frequency, you over represent/cahnge ratio. instead of maintain ratio, you change it.
define purposive sampling
biased sampling where only specific kinds of people are included, targeting very specific group of people
define snowball sampling
baised sampling technique in which particpatns are asked to recommend other for the study. asking them to see if a friend would do it, can create bias as those people are more likely to be similiare, not gonna perfectly represent whole popluation.
define quota sampling
where researchers identify subsets of the population, set a target number for each category and nonrandomly select individuals within each category until the quota is filled