Task 5 Are you sure? Flashcards
External validity
Its results can be extended beyond the limited research setting and sample in which they were obtained
• Should be conducted in a way that they can be used in the real-world
• gain insight in underlying behaviour
Internal validity
- Internal validity is the ability of your design to test the hypothesis that it was designed to test
- Testing your hypothesis
- Showing that only the independent variable causes the observed variation
- Are threatened by rival hypothesis
Confounding
Two ore more variables are combined in a way that you can’t see which observations are from which variable
Construct validity
- Convergent (similar to instruments measuring the same construct)
- Divergent (not similar to measure of a different construct)
Pretesting
When you have to test your participants if they fulfil your requirements. Can lead to less generalization (threat to validity)
Biased sources
Unrepresentative sources (e.g. asking republicans if weapons should be forbidden) (threat to validity)
Sampling error
When your sample differs in the characteristics compared to your population (threat to validity)
Volunteer bias
volunteers differ in meaningful ways from the general population, the difference between volunteer and nonvolunteer affect the external validity of your research.
Generalization
the ability to apply findings from your sample to the whole or larger population
Random sample
Pick random people out of your population
Non-random sample
usually individuals from a highly specialised suppopulation (e.g. college students)
→less external validity
Probability sample
Everybody from your population has ideally the same probability to get in your sample
1. Representativeness:
→Sample should be representative for your population
→biased sample when its not representative
random sampling
everybody has an equal chance to appear in your study
simple random sampling
choose a random sample e.g by using a random number generator
→ random digit dialling: use 4 numbers and one exchange digit to get the random factor
Stratified sampling
you divide the population into segments and then you take equally large samples from each segment