Lecture 3 Flashcards
Define Validity
The extent to which a procedure measures what it is intended to measure
What are the two types of validity and there definitions
Internal - Ability to say the predicted relationship between the IV and DV does exist and is due to the effects of the IV, not EVs
External - degree to which the reulsts can be generalised to other subjects/ situations.
Define secondary variance
effects of uncontrolled variabes
What are the 3 ways of dealing with secondary variance?
1) Record it and correct for it in stats
2) Eliminate it
3) Hold it constant
What affect does increased internal validity have on external validity?
Higher internal may compromise external
What factors can affect validity?
- multiple treatment affectes
- reactive effects
- demand characteristics
- differential treatment of each p
- different time/ place
- participant drop outs/ low response rates
- regression towards the mean
- order effects
- maturation
- non-random assingment
- time between multiple observations
Define observational studies
- often the background for intervention designs, to determine if one factor is related to another factor.
- Observe natural variation in IV and DV, testing the correlation between both, controlling for other known determinants of DV
Define intervention studies
only way to approach cause and effect – control all variables except IV to see the effect on the DV
Define and Evaluate Questionnaires
Usuaully paper test, score answers to draw conclusions √Predictive and very useful if useful.
X – honesty is questionable
Define and evaluate survey’s
Either written questionnaire, verbal interview or both
√Gather info on things that aren’t easily observable
√Gather lots of data quickly
X - honesty is again questionable
Define and evaluat case studies
Involes interviews, observations, examine records and psychological testing to see what evental led up to the current situation. It is an in-depth investigation of an individual, used to reconstruct parts of their life.
√ - Good for testing psychological disorders as you can see history and development
X - Very subjective, with their being often gaps and relying on memory of subject, which may be faulty
X - cant always generalise
What are the two types of observations?
Lab and naturalistic
Define and evaluate lab observations
In a controlled environment, usually with an observation schedule.
√ More structured
X - observer effects
X - low external, cant generalise - ecological validity
Define and evaluate naturalistic observations
In a natural environment, allowing behaviour to occur without interference/ intervention by researcher (we all do this i.e. people watching).
√ - No observer effects
X - intrusive/ ethics
Define reliability
The extent to which a particular measure is consistent and reproducible. Measures include: questionnaires, tests, a computer that records responses, a stop watch. Reliability depends on measurement error. If something is reliable it means the test will yield comparable results when repeated measures are taken under matched conditions.
