Validity

• Validity 


  A test is valid when is measures what it claims to be measurin
• Two ways of assessing validity

  .Face validity

  .Concurrent validity

  .predictive validity
• Face Validity


  whether As test appears to measure what it claims to
• Concurrent validity


  Comparing a results to a previous test by another test of the same behaviour that is valid to see if the first test has ‘concurrent validity’. Of correlation between the two tests results is over 0.8+ it has high concurrent validity
• Internal validity

  .whether researcher measured what they intended to measure.
• External validity


  Whether findings can be Generalised beyond study and researcher settings
• internal validity- within tests

  .social desirability- participants hide their genuine opinions/behaviours to look good and be more socially acceptable

  Demand characteristics-participants who has discovered the aim and try support results

  .investigator effects-researcher bias where the researcher un/consciously influences results

  .uncontrolled extraneous variables-lack of controls ;not using standardised procedures not controlling participant variables by randomly assigning to groups
• External validity- outside tests

  Ecological validity-can be generalised to alternative enviorment( test of obedience from lab studies to a busy street ,could obedience be replicated)

  .population validity- sample of a study is representative of target population (gender ,ages ,ethnicity)

  .temporal validity- findings can be generalised to other time periods .tipics like social influence ,attachments ,relationships would be the same in modern society
• predictive validity


  .the extent to which the test can predict future performance/outcome and behaviour (GCSE scored are highly predictive of A-LEVEL scores)
• Improving validity- Internal


  .random allocation- controls participant variables

  .standardised procedures- controls extraneous variables

  .single and double blind trials- controls research bias+ demand characteristic

  .peer review-control research bias.
• Improving vallidty-external

  .replicating in multiple settings- improves ecological validty

  .replicating with diverse groups-improves population validty

  .replicating historical studies-improves temporal validity .

  .replicating findings with real world tasks
• Mundane realism - external validity

  These studies simulate everyday scenarios making it easier to apply to findings to real life behaviour or outcomes instead of a laborstory