Research methods- Sampling Flashcards
What are the 5 types of sampling?
-Stratified
-Random
-Systematic
-Volunteering
-Opportunity
(from most representative to least representative)
What is an opportunity sample?
-Sample is already there.
-e.g. students in a school.
Strength (O)
Quick and easy
-Make use of people closest
-Cheaper
-Less time-consuming
Limitation (O)
Unrepresentative/ lacks generalisation
-Drawn for one area
-So data cannot be generalised
-Lacking external validity
What is a Volunteering sample?
-Participants volunteer to take part.
-e.g. Milgram.
Strength (V)
Easy
-Requires minimal input from the researcher
-Less time-consuming/ effort to find participants
-Participants select themselves
Limitation (V)
Volunteer bias
-May attract certain profile of person, then may affect generalisation
-Responds to cues
Demand characteristics
-Reduces validity if know aims from the advert, so may act unnaturally
What is a Systematic sample?
-Use a system to choose participants.
-e.g. names into hat/ generator= every 5th person.
Strength (Sy)
Unbiased selection
-Avoids researcher bias/ Researcher no influence
-First item usually selected at random
-Objective
Limitation (Sy)
Only possible if whole target population can be identified
-Time/effort
-Complete list of target population is required
-May as well use random sampling
What is a Random sample?
-Putting names into a hat.
-e.g. all psychology students, first 20 make the sample- equal chance of being selected.
Strength (R)
Unbiased selection
-CVs/EVs are controlled
-Enhances internal validity
-Stops researcher bias as they have no control over who is selected so they cannot pick people who will follow their hypothesis
Limitation (R)
Impractical
-Time-consuming
-A complete list of TP is difficult to obtain
May not be representative
-Possibly could select all female participants for the sample
What is a Stratified sample?
-Work out % of people (subgroups) that make up population, get Strata from them, use random afterwards.
-e.g. how many men/women, work out percentage over 100.
Strength (St)
Unbiased/ Representative
-Avoids researcher bias
-Once subdivided, random selection occur
-Designed to reflect target population
-Can generalise
-High external validity