sampling methods Flashcards
what is probability based sampling
random sampling - ppts have equal likelihood of being picked
what is non probability sampling
convenience sampling - probability not known or left up to the individual to choose
simple random sampling
any two groups of equal size, equally likely to be selected
stratified random
stratify population into
homogeneous groups then use simple random sampling to
draw samples from each group
cluster sampling
when natural sampling unit is a
group/cluster of individual units, sampling within
group/cluster (
systematic sampling
election of every kth element from
a sampling frame or from a sequential stream of potential
respondents
quota sampling
specify desired number of
respondents with certain characteristics
snowball sampling
relies on referrals from initial
respondents
judgment sampling
researcher selects sample
based on their judgement
population of inference
population about which the researcher intends to draw conclusions
target population
population of inference minus any groups the researcher has chosen to disregard
frame population
portion of the target population to which the survey materials/devices delimit,
identify, and subsequently allow access
what is post stratifying
(weight survey sample to match population of inference on observed key
characteristics)
coverage error
failure ot give any chance of selection to some persons in the pop
sampling error
heterogeneity on the survey measure amongst the pop
nonresponse error
failure to collect data on all persons in the sample
measurement error
innaccuracies in responses recorded
where may low quality data reside from
boredom, distraction, fatigue and random response
issue with online study
interent users limit generalisability
what is self selection bias
participants with a personal interest in a
topic may be more interested to take part in a study on that
topic
advantages of online collection
- Faster response times
- Increased response rates
- Error messages
- Can incorporate links
- Difficult skip patterns
- (Relatively) easy to design/produce.
- Large samples can be obtained.
- Can reach participants who would not attend a lab session, who are far away geographically etc.
- Decreased costs (monetary and environmental)
- High levels of measurement equivalence between computer-based and paper-and-pencil formats
issues of online studies
Requires internet access
* Social desirability bias
* Incomplete data
* Data can be of low quality (e.g., due to participant fatigue/boredom, random
responding etc.)
* Lack of control over environment/situational factors
* Security/access, privacy, and ethics
* Sampling biases (samples may not be representative of the general
population etc.)
