sampling techniques Flashcards
sample
the people the sociologists actually study in their research
sampling frame
a list of all the people in the population who are relevant to the particular topic being studied
random sampling
random sampling - purely random using eg random number generators
systematic sampling - where you select every nth name from a list
random sampling advantages
- ensures every individual has absolutely equal chance of being selected
- quick and easy to carry out
random sampling disadvantages
- could be unrepresentative eg by chance more males than females
- requires accurate, up-to-date and complete sampling frame
stratified random sampling
to stratify your sample so it represents the general population - you then draw randomly
stratified random sampling advantages
- ensures control of variables such as age and sex that are important to the research
- can increase precision and so improve representativeness of sample
stratified random sampling disadvantages
- complex and time-consuming to identity to strata and to sample various sub-groups
- sampling frame may not contain required info to split sample-population into groups
- your values and views on society may influence choice of stratification - bias
cluster sampling
the entire population is divided into groups (cluster) and a random sample of these clusters is selected
all observations in the selected clusters are included in the sample
cluster sampling advantages
- useful if a complete list of the population is unavailable, but you can get a complete list of groups or clusters of the population
- cheaper than random sample if list of subjects is widely scattered
cluster sampling disadvantages
could be unrepresentative as clusters may not accurately reflect the general population
quota-sampling - non
find a specific number of a particular types of person, once quota filled they stop
often in market research
quota sampling advantages
- can control variables without a sampling frame
- cheap and quick
quota sampling disadvantages
- not random
- may have to ask personal questions to discover whether respondent matches the necessary criteria of study
- not representative as quota sampling
snowball sampling - non
respondents recruit one individual who will then get people they know to participate in the study - these people will then get further participants