reliability

• what is reliability

  the consistency of measurements. we would expect any measurement to produce the same data every time it is used
• 2 ways of assessing reliability

  test-retest

  inter-observer reliability, inter-interviewer reliability (for interviews), inter-rater reliability (for content analysis)
• test-retest

  the same assessment given to same participants on 2 occasions to see id the same results are obtained. results are correlated and if they have correlation coefficient of +0.8 or more= high test-retest reliability
• inter-observer reliability

  extent to which there is agreement between 2 or more observers involved in observations of behaviour. if scores have correlation coefficient of+0.8 or more= good inter observer reliabilty
• correlation assessment for questionare

  • each ptpts score from first questionnaire is correlated to second questionnaire scores

  •shown in scattergraph- first test scored are plotted on one axis and the second test scores in the ither

  •strength if correlation is assessed using spearmans rho test or pearsons r test
• improving reliability of observation

  ensure behavioural categories are operationalised clearly- no overlapping and are measurable
• improve reliability on questionares

  ensure questions are not too complex- replace open, ambiguous questions to close, fixed choice alternatives
• improve reliability in interviews

  use the same interviewer each time if possible. if not then ensure interviewers are properly trained and use more structured interviews
• improve reliability in experiments

  ensure procedures are standardised