Reliability

• Reliability

  Refers to how consistent a measuring device is, this includes psychological tests or observations which assess behaviour.
• Test-retest reliability

  A method of assessing the reliability of a questionnaire or psychological test by assessing the same person on 2 separate occasions.

  Shows the extent to which the test produces the same measure.

  • Involves administering the test or questionnaire to the same person on 2 separate occasions.
  • If test is reliable then results obtained should the same or similar each time.
• Why should time be left in between retests in test-retest reliability?

  Must be sufficient time between test and retest to ensure participants cannot recall their answers but not so long that their attitudes, opinions and abilities change.
• Inter-observer reliability

  • The extent to which there is an agreement between 2 or more observers involved in an observation of behaviour.
  • This is measured by correlating observations of 2 or more observers.
  • +80 = High inter-observer reliability.
• Improving reliability in questionnaires

  • Test-retest method
  • Comparing 2 sets of data should produce a correlation that exceeds +80.
  • A questionnaire that produces low test-retest reliability may require some of the items to be deselected or rewritten.
  • Replacing open questions which closed fixed-choice alternatives which may be less ambiguous.
• Why would questions need to be removed from a questionnaire?

  • If some questions are complex or ambiguous they may interpretative differently by the same person on 2 separate occasions.
• How to improve reliability of interviews?

  • Use the same interviewer every time.
  • If this is not practical, all interviewers should be trained.
  • More easily avoided in structured interviews where the interviewers behaviour is controlled by fixed questions.
  • Unstructured interviews are more free flowing and less reliable.
• Improving reliability of experiments?

  • Procedures that focus of reliability.
  • In order to compare performance of different participants the procedure must be the same every time (standardised procedures)