Vaildity

• what is validity

  extent to which a result is genuine (legit)
• what is internal validity

  whether its only the IV that affects the DV
• whats external validity

  whether results can be generalised
• 3 types of external validity

  ecological, population & temporal
• whats ecological validity

  whether results can be generalised to other settings - e.g. 'everyday life
• whats population validity

  whether findings can be generalised to other groups of people
• whats temporal validity

  whether findings can be generalised over time
• whats face validity


  whether findings appear 'on the face of it's to measure what it claims to be
• how do you determine face validity


  getting an expert to check, ask others whether its a good measure, scrutinise the measuring instrument
• whats concurrent validity

  when results are very close/match the results from another recognised/well established test
• how do you establish concurrent validity

  - correlate 2 sets of scores

  - if correlation coefficient >0.8 = significant
• what would indicate high concurrent validity


  close agreement between 2 sets of data
• how do you improve experiments validity (4)

  - standardisation

  - control groups

  - single/double blind procedures
• how do you improve validity in questionnaires (3)

  - remove problematic questions (ambiguous, leading, double-barrelled)

  - include a lie scale

  - assure Ps data will be anonymous
• how do you improve observations validity (2)

  - behaviour categories (non-ambiguous, not too broad)

  - higher ecological validity if covert
• how do you improve validity when using qualitative methods (3)

  - triangulation (use number of different sources for evidence)

  - interpretive validity of conclusions (do Rs/Ps interpretations of events match)

  - higher ecological validity (depth/detail of case studies reflect Ps reality)