Validity

• Validity is when findings accurately reflect natural behaviours and factors.
• Internal validity questions the cause and effect relationships between the change made to the IV and observed change to the DV. If the change to the DV is influenced by any other factor, then the findings lack internal validity.
• External validity is when findings are able to be generalised beyond the study.
• Internal validity should consider:

  • Social desirability,
  • Demand characteristics,
  • Investigator effects,
  • Uncontrolled extraneous variables.
• External validity should consider:

  • Ecological validity,
  • Mundane realism,
  • Population validity,
  • Temporal validity.
• Assessing validity:

  • Face validity - Does the test appear to measure what it claims to be measuring?
  • Criterion validity is the confidence in the validity of a test increases if we can compare the data from a test to another measure of the same variable and identify a correlation.
• Concurrent validity is the extent to which data from a newly created test is similar to an established test of the same variable conducted at the same time. (Correlation = Strong is it is +0.8 or higher).
• Predictive validity is the extent to which a performance on a test can predict future performance, outcomes or behaviour.
• Improving internal validity:

  • Random allocation,
  • Standard procedures,
  • Counter balancing,
  • Single and double blind trails,
  • Peer review.
• Improving external validity:

  • Multiple settings to improve ecological validity,
  • Diverse group,
  • Replicate historical findings,
  • Realistic tasks.