838222
| Question | Answer |
| Sampling | Who is going to participate |
| Issue of external validity? | To what degree can one generalise the findings from sample to population? |
| Representative | It is important that the Sample be Representative of the Population |
| Sampling Frame | The actual part of the population from which I draw my sample |
| Sample | The subset of the sampling frame from which I obtain data |
| Probability Sampling | Expensive, time consuming, better than non-probability sampling |
| Single Random Sampling | Every person in the population has an equal chance of being sampled. From a list of all members, but sometimes some members are not included, e.g., home school children. |
| Stratified Random Sampling | Divide population along dimensions (e.g., gender) and be sure that you sample sufficiently across these dimensions. Can Oversample for important dimensions, e.g., ethnicity |
| Cluster Sampling | Obtain participants fom pre-existing groups or clusters. E.g., Children from NZ schools |
| Non-probability Sampling | Cheap, easier, but not representative |
| Haphazard Sampling | Convenience sampling, handy for the researcher, but biases are introduced. |
| Quota Sampling | Obtain appropriate percentages of different types of participants (e.g., gender, etc.), but one is still obtaining these participants from readily available sources so not representative but more so. Also introduces biases |
| Purposive Sampling | You select individuals who fit within a particular category to fit a purpose. E.g., testing/sampling from people who have previously had depression |
| Most common and why? | Most psychologists use haphazard non-probability sampling. They only want to explore the relationships between variables. It is much cheaper |
| Biases | If sample frame is hetrogeneous with population then biases are introduced. It is non-representative |
| Creative approaches to collecting data and encouraging compliance | Passive consent (please tell us if you don't want to participate), but this can be controversial. Compensation and Inducements. Interesting ways to collect data; laptops; internet; Ipads. The "personal touch" using interviews (time consuming, expensive). Underutilised Samples (kids in Camps, kids in malls) |