Reliability means consistency. The results of a research study are reliable if, when the study is replicated, the same results are consistently found.

This replication involves repeating a research study under exactly the SAME CONDITIONS, using exactly the SAME PARTICIPANTS.

Be aware: The results of an investigation may be reliable, but that does not mean that they are valid. A study can be reliably invalid (i.e. consistently produce inaccurate results).

Issues of Reliability (A Level Only)

Psychologists need to measure variables consistently. This aspect of research is known as reliability. If you are a reliable student you regularly turn up to your lessons and always hand in work on time. A reliable piece of research should always produce the same or similar results when replicated in exactly the same conditions with the same participants. Reliability is used to assess both experimental procedures and ‘tools’ such as tests, questionnaires, interviews and behavioural categories in observations.

Types of Reliability:

There are two types of reliability:

(1) Internal Reliability (the consistency of the measure within itself)

Internal reliability refers to the consistency of a measure within itself. Internal reliability refers to the consistency of a measure within itself. For example, the items on a questionnaire or questions in an interview should be testing the same thing.

(2) External Reliability (the consistency of a procedure from one occasion to another)

External reliability refers to the consistency of a procedure from one occasion to another. For example, an experiment performed on two different days, in different laboratories, or by different researchers should still produce similar results (e.g. two researchers using the same interview format, equipment, behavior schedule or test should obtain the same results).


Assessment of Reliability (A Level Only)

The two types of reliability can be tested separately to assess levels of internal reliability as well as levels of external reliability. Just because a test has high internal reliability does not therefore mean it will also have high external reliability. For this reason, it is worthwhile assessing both types.

Internal reliability can be assessed by:

1. Split-half reliability:

If you measure someone’s IQ today you would expect to get a similar result if you used the same test to assess the same person in a few weeks’ time. If the results were the same time (i.e. if the results were consistent, you could assume the test was reliable). Rather than waiting a few weeks to try the test again it is possible to use split test reliability. For example with an IQ test, split it in half give both halves to the participant and compare their score on each separate half. If scores on each half are similar psychologists assume the test to be reliable.

2. Equivalent forms reliability:

Two tests, questionnaires or structured interviews of the same type are given to the same participants. Participants’ results on the two forms of the test should correlate strongly if the tests are reliable.

External reliability can be assessed by:

1. Test-retest reliability: Participants take the same test twice, at different times. If the results for the two occasions correlate, the test has high external reliability. This can identify individual items that generate inconsistent results or other factors that cause variation, such as different settings or researchers. This is typically used to test reliability of structured interviews or questionnaires.

2. Inter-rater/observer reliability: Two (or more) observers watch the same behavioural sequence (e.g. on video), equipped with the same behavioural categories (on a behavior schedule) to assess whether or not they achieve identical records. Although this is usually used for observations, a similar process can be used to assess the reliability of interviewers.