generalization
= apply findings from a sample to lager population
–> studies who’s findings can be applied across a variety of research & subjects populations processes a high degree of external validity
sample
- -> small subgroup choose from lager population
- need sample of population (normally not possible to study entire population)
- sometimes also subpopulation (costs, other factors max limit you to study certain region)
(simple) random sampling
= if everyone has an equal chance pf being chosen for the study
–> pick randomly participants form population to take pace in your study
(still does not guarantee random)
nonrandom samples
- not easy to really get random people for study to be able to apply findings to a whole generation
- -> often just college students for study
e. g. internet study –> only people who can work on the internet/ animal subjects
laboratory research
= bringing participants into laboratory environment of your creation (experimenter as a lot of control)
field research
= conduct your research in participants neutral environment (less control of researcher)
- must select participants while in their natural environment
volunteer bias
- people who volunteer for a study also not presentative for whole population
- -> experienced as more open, educated, more interested in the study,..
systematic sampling
pick randomly e.g. every 4th person from your list of participants
stratified sampling
divide population in characteristics and take same amount of participants from each group (e.g. 5 black, 5 white..)
proportionate sampling
dividi population in groups & calculate % of these groups in population
- -> take same % of participants of different group in your study
(e. g. in population 60% asian –> in experiment 60% are asian)
cluster samling
= naturally occurring groups of subjects and randomly selecting certain clusters
–> multistage sampling: selecting individuals from those clusters
basic sampling ideas
- representative sample
- non-representative sampling
- random sampling
- non-random sampling
- matches characteristics of the population
- biased
- identify the population and then draw a random sample from it
- choosing special individuals/ subject pools
e. g. only university students/ internet research/ animal subjects (same breed, sex etc.)
kinds of sampling
- simple random sampling
- stratified sampling
- proportionate sampling
- systematic sampling
- cluster sampling
- randomly selecting a number of people
- dividing population into segments (=strata), then selecting a sample of equal size from each segment
- proportions of people in the population are equal to proportions in your sample
- every Kth element after a random start
- naturally occurring groups of subjects and randomly selecting certain clusters
- Multistage sampling: selecting individuals from those cluster
