reliability and validity

test-retest used for testing reliability
administering the same test/questionaire to the same person on different occasions
if reliable the results should be similar if not the same each time it’s administered (correlation of +0.8 or higher) - if good then its reliable
must be a significant time between retesting so that participants cannot recall answers given previously
not so long that opinions may have changed

inter-observer reliability used to test reliability
2 observers of the same study
a small-scale pilot study to check that observer is applying behavioural categories in the same way
data is collected by both observers and compared to see correlation if reliable

validity is the extent to which an observed effect is genuine (can it be generalised to rest of the public)
face validity - does it measure what it is meant to measure
concurrent validity - extent to which a psychological measure relates to an existing similar measure
ecological validity - can findings be generalised to other settings/situations (form of external validity)
temporal validity can findings be generalised to other historical times (form of external validity)