Reliability Across Methods

• What is reliability in the context of investigation methods?

  Reliability is a measure of how consistent the findings from an investigation are.
• Why is reliability important in investigations?

  It ensures the dependent variable is measured accurately and conclusions are valid.
• What can happen if reliability is not ensured in research?

  It can have implications for theory development.
• What are the types of reliability assessment?

  • Internal reliability
  • External reliability
• What does internal reliability describe?

  It describes how consistent something is within itself.
• What is the split-half method used for?

  It assesses internal reliability by comparing two forms of the same test.
• What correlation coefficient indicates high internal reliability in the split-half method?

  ≥ 0.80
• What is external reliability?

  It is when consistent results are produced regardless of when or who administers the investigation.
• How is the test-retest method used to assess external reliability?

  The same test is administered to the same person on different occasions.
• What correlation coefficient indicates high external reliability in the test-retest method?

  ≥ 0.80
• What is inter-observer reliability?

  It refers to the extent of agreement between two or more observers observing behavior.
• How is inter-observer reliability calculated?

  By the formula: Total number of agreements / Total number of observations.
• What correlation coefficient indicates high inter-observer reliability?

  ≥ 0.80
• How can reliability be improved in questionnaires?

  • Deselect or rewrite items with low test-retest reliability
  • Replace open questions with closed, fixed-choice alternatives
• What are the best practices for ensuring reliability in interviews?

  • Use the same interviewer each time
  • Train all interviewers properly
  • Structure interviews in a consistent manner
• Why are lab experiments often considered reliable?

  Researchers can exert strict control over many aspects of the procedure.
• What is a challenge of conducting field experiments compared to lab experiments?

  Variables are much more difficult to control in the field.
• How can the reliability of observations be improved?

  • Properly operationalize behavioral categories
  • Ensure categories are measurable and self-evident
  • Avoid overlapping categories
• Why is it important to operationalize behavioral categories well?

  To avoid differing and inconsistent records from different observers.
• What is an example of a well-defined behavioral category?

  'Pushing'
• What is an example of a poorly defined behavioral category?

  'Aggression'
• What should be avoided in behavioral categories to improve reliability?

  • Overlapping categories
  • Absence of categories