Validity

• Validity refers to the extent to which the observed effect is genuine
• Internal validity refers to whether the effects observed in an experiment are due to the manipulation of the IV and not other variables
• Threats to internal validity:

  • Extraneous and confounding variables
  • Investigator bias
  • Demand characteristics
  • Social desirability bias
• Extraneous and confounding variables- e.g. the time of day, noise, temperature could affect the DV if not controlled
• Investigator bias- researcher could have influenced the participants in some kind of way
• Demand characteristics- participants may have guessed the aim of the study and changed their behaviour
• Social desirability bias- participants change their answer or behaviour to make themselves look better
• External validity refers to a measure of whether data can be generalised to other situations outside of the research environment
• Types of external validity:

  • Ecological validity
  • Population validity
  • Temporal validity
• Ecological validity- whether findings are generalisable to the real world
• Population validity- how well findings can be applied to a population as a whole
• Temporal validity- how successfully research findings can be applied across time periods
• Ways of assessing validity:

  • Face validity
  • Concurrent validity
• Face validity:

  • a measure is scrutinised to determine whether it appears to measure what it is supposed to
  • done by examining the measuring tool closely or having an expert examine it
• Concurrent validity:

  • whether research findings demonstrated are close to or match those obtained on another recognised and well-established test
  • checked by comparing a participant's score on your measure with their scores on an existing well-established measure
  • close agreement between the two data sets indicates high concurrent validity
• Improving validity on questionaires:

  • incorporate a lie scale to assess consistency of ppts answers, control effects of social desirability and acquiescence bias
  • anonymous
• Improving validity in qualitative methods:

  • triangulation: use different methods of data collection
  • interpretative validity: use direct quotes
• Improving validity in experiments:

  • control groups: whether DV causes the change in IV
  • Standardise procedures: minimises investigator effects and demand characteristics
  • single-blind and double-blind procedures
• Improving validity in observations:

  • covert: behaviours are more likely to be authentic
  • clear and operationalised behavioural categories