reliability

Question 1

reliability

Answer 1

Used to assess both experimental procedures and ‘tools’ in tests, interviews and behavioural categories. A reliable psychological test will always produce the same results.
Reliability means consistency - something that can be depended on.
A study is reliable if it can be replicated and the results are similar each time.

Question 2

internal reliability

Answer 2

Consistency within itself e.g. a questionnaire should all measure the same personality/ trait.
For example, Eysenck’s personality questionnaire.

Question 3

external reliability

Answer 3

Consistency through different occasions, researchers and laboratories.
For example, Milgram’s research.

Question 4

test-retest

Answer 4

Assesses reliability of questionnaires or psychological tests (e.g. IQ tests).
Involves giving the same test or questionnaire to the same person on different occasions.
If the results are similar then we can say it is a reliable measure.

Question 5

inter-observer reliability

Answer 5

If the study involves several observers working together to observe different participants, it is essential to assess whether they are all interpreting the behaviour the same way. This can be done by:
- making sure the operational definitions of behaviour are clear.
- training the observers thoroughly.
- recording behaviour to check what actually went on.
- correlating observers’ results.

Question 6

improving reliability

Answer 6

Questionnaires - measure over time using test-retest.
Interviews - use the same interviewer each time, train the interviewer, avoid ambiguous/leading questions.
Experiments - use standardised procedures, laboratory rather than field for control, test under same conditions.
Observations - operationalised behavioural categories, categories shouldn’t overlap, all possible behaviours should be on checklist.