reliability Flashcards
What is reliability in psychological research?
Reliability refers to the consistency of a measure or procedure—i.e., the extent to which it produces the same results under the same conditions.
What is internal reliability?
Internal reliability refers to how consistently a measure assesses something within itself (e.g., all items in a questionnaire should measure the same construct).
What is external reliability?
External reliability is the extent to which a measure or procedure produces consistent results over time (e.g., retesting should yield similar scores).
What is the test-retest method and what does it measure?
The test-retest method involves giving the same test to the same participants on two different occasions and comparing the scores. It assesses external reliability.
What is inter-observer reliability?
This assesses how much agreement there is between two or more observers coding behaviour in an observation. High agreement = high reliability.
How is inter-observer reliability calculated?
By comparing the observers’ scores or tallies using a correlation coefficient (usually aiming for +0.8 or higher for good reliability).
How can internal reliability be assessed?
Using the split-half method: a test is divided into two halves and the scores are compared. High correlation indicates high internal reliability.
What does the term “reliable measure” mean in the context of psychological testing?
It means the test or tool gives stable and consistent results, and any variation is due to true differences—not measurement error.
How can reliability be improved in experiments?
• Use standardised procedures
• Ensure clear operationalisation of variables
• Use objective measures where possible
How can reliability be improved in observations?
• Use a clear coding system
• Train observers thoroughly
• Conduct a pilot study
• Assess inter-observer reliability
How can reliability be improved in questionnaires?
• Remove or revise ambiguous or leading questions
• Use closed questions with fixed response categories
• Conduct a pilot study
What do examiner reports say about common errors with reliability?
• Students confuse reliability and validity
• Vague answers like “do it again” instead of explaining test-retest
• Not specifying how reliability would be measured or improved
• Confusing inter-observer reliability with inter-rater validity
How can you use reliability to evaluate psychological research in AO3?
You can critique a study by asking:
• Were the procedures standardised and replicable?
• Were the measures used reliable over time or between raters?
• If not, the findings may lack consistency and trustworthiness
