reliability

• internal reliability= consistency inside the study
• external reliability = whether findings are consistent outside the study
• internal reliability issues:

  • more the one researcher
  • more then one measure
  • within a measure
• more then one researcher= is there consistency in interpretations
• more then one measure= is there more then one test and is there consistency of challenge
• within a measure= if a measure has more then one question or item and are they consistently measuring the same thing
• to deal with more then one researcher= clealry operationalise variables and standardise procedures
• to deal with more then one measure= standardise the measures
• to deal within the measure= standardise each item or question in the study
• external reliability issues:

  • behaviour is only measured once
• behaviour is only measured once= the test may not measure behaviour consistently at other times
• to deal with that behaviour is only measured once= ensure tests are designed in relation to measuring behaviours
• how to assess reliability:

  • inter-rater reliability (internal)
  • spilt-half (internal)
  • test-retest (external)
• inter-rater reliability= two or more psychologists produce consistent results and if there is an 80% or above agreement
• split half= splitting the ppts test answers in half and seeing if they got similar scores on the two halves
• test retest= testing and retesting the same ptts over time with the same test and comparing the results