Sampling Techniques Flashcards
(15 cards)
Random
Each member of target population has equal chance of being selected
Need a list of target population then randomly draw names out of a hat or name generator
Random - strength
Unbiased selection- cv/EV are controlled, enhances internal validity
Random weakness
Impractical- complete list of population is hard, some participants may may refuse to take party
May not be representative by chance - still end up with sample that’s not representative, may get 20 females in one and 5 in another
Opportunity
Involves using participants who are available and willing o take part - asking people in the street
Opportunity - strength
Quick and easy - make use of people who are closest - cheaper, popular
Opportunity - limitation
Unrepresentative/acts generalisation - unrepresentative of target population - drawn from specific area - findings no generalised
Volunteer
Participants select themselves to take part (advert or ask participants to take part)
Volunteer - strengths
Easy - select themselves and are willing, less work for researcher, engage more
Volunteer - limitation
Volunteer bias - share certain traits (want to be helpful), not generalised
demand characteristics
Systematic
Where every nth member of the target population is selected
Need a list of everyone in target population
Systematic - strengths
Unbiased selection - first item usually selected at random - objective method
Fairly representative- would be possible but unlikely to get sample o target populaton
Systematic - limitation
Only possible if you can identity all member of target population - complete list is required to work, may aswell use random sampling
Stratified
Makeup of sample reflects proportions of people i certain subgroups within population
Must identify subgroups then work out proportion to make representative
Then subgroups picked from Radom sampling
Stratified - strength
Unbiased/representative - characteristics representative and generalised more than other methods
Stratified - limitation
Time consuming - knowledge fo characteristics required, takes time to work out subgroups and them mak sample
Stratified not perfect - have to reflect all ways in which people are different. Complete representation not possible