reliability

• reliability

  • if a particular measurement is repeated and the same result is obtained then tat measurement is reliable
• ways to asses

  • test-retest - the same person is given the same test on two or more occasions. if the test is reliable results should be the same or very similar
  • inter-observer - 2 or more observers agree the behavioural categories. sit and record separately. In both tests correlations are used and the coefficient must be +0.80 for reliabilitiy
• ways to improve

  • questionnaires - rewrite qs (replaced open for closed, fixed qs - less ambigious)
  • interviews - improve training - always try use the same interviewer, use a trained interviewer so they avoid leading/ ambiguous qs
  • observation - behaviour categories must be operationalised, must not overlap
  • experiments - procedure must be the same way for everyone - every ppt does the same thing in the same order