Sampling

Cards (6)

  • Sampling
    • it's not possible for everyone to take part in an experiment. Psychologists therefore have to select a sample from the target population (a group of people who researchers want to apply their findings to)
    • the sample should represent the target population so the findings can be applied to them
    • Sampling bias may occur if the sample is not representative. To avoid this, the sample should be as large as possible
  • Volunteer sampling
    participants are self-selecting (they choose to take part in research). This is done through advertising - such as posters
    S - volunteer sampling is not time-consuming, it is also cheap
    W - all participants are likely to be similar (extroverts)
  • Opportunity sampling
    using whoever is avaliable at the time (stopping people on the street)
    S - not time-consuming. Participants are easily accessible and gathering the sample is also cheap
    W - likely to produce a bias sample as participants will only be selected from limited areas
  • Random sampling
    every member of the target population has an equal chance of being chosen. This can be done by drawing numbers from a hat (put all the target population in a hat, blindfold a researcher and get them to pick a name out, shuffle and repeat) or a random number generator (assign a number to each member of the target population, get an independent researcher to draw numbers from number generator, match the numbers to names)
    S - representative as findings can be generalised
    W - could be impractical if the target population is large
  • Systematic sampling
    sampling in a systematic way (taking every Nth person from a list)
    S - no researcher bias
    W - a complete list of the sampling frame might be unavailable
  • Stratified sampling
    you need to identify the sub-groups in a target population. Calculate the proportions needed for each group. Randomly select participants from each group and assign to experimental conditions. The percentage in the sample is equal to the percentage in the target population (proportionate)
    S - findings can be generalised as it's representative
    W - it can be difficult and time-consuming to put participants into sub-groups