reliability

• reliability refers to how consistent a measuring device is inc psychological tests. psychologists tend to measure abstract concepts such as aggression through interviews and questionnaire
• test-retest:

  involves administrating same test or questionnaire to the same person on different occassions . if it is reliable then the results will be the same or similar . must be a sufficient time between the test so pps cannot recall previous answers but not so long that their attitudes and opinions change. correlated above +.8
• inter-observer reliability:

  observers should do a pilot study to make sure that they are applying the same target behaviours or a comparison at the end of the study. observers records data independently. corrlate +.8
• improve reliability questionnaire:

  may require that some of the items are deselected or rewritten especially if wording complex or ambigious as interpretated differently by others
• improve reliability interviews:

  use same interviewer each time if not possible all interviewers must be trained to the same standard

  use structured interview

  no leading questions
• improve reliability observations:

  behavioural categories fully operationalised and measurable so that they do not overlap. observers may make their own decision therefore inconsistent records

  if low then more training behavioural categories so they are more consistent
• improve reliability experiments:

  compare the performance of different pps well produces must be standardised