AQA Psychology Reliability/Validity

Internal reliability

The extent to which something is consistent within itself. For example all the questions on a IQ test are measuring the same thing

External reliability

Measure of consistency over several different occasions. The outcome should be the same if an interview takes place a day later, a week later

What are the two ways of assessing reliability

Test retest reliability

Inter-observer reliability

Test retest reliability

The same test is administered to the same person or group on different occasions and the results are compared

Inter-observer reliability

Agreeing beforehand on behavioural categories to observe the behaviour and tally independently tallies correlated using a statistical test.

How can you improve reliability in a questionnaire

If a questionnaire has a low test retest some question's need to be less ambiguous or more closed questions

How can you improve reliability in a interview

Make sure interviewers are trained.

Avoid leading and/or ambiguous questions

How can you improve reliability in an experiment

Standardisation of procedures will minimise extraneous variables

How can you improve reliability in an observation

Behavioural categories should be properly operationalised and not overlap

What are the two types of validity?

Internal and external validity

Internal validity

How accurately a test or measuring instrument measures what it says it measures

What are the two ways of assessing internal validity

Concurrent and face validity

Concurrent validity

Can be established by comparing performance on a new questionnaire or test with a previous questionnaire or test. If participants performance show a high correlation this is high evidence of concurrent validity

Face validity

Less rigorous and involvesee looking at the questions to see if they are genuinely about what they should be asking

External validity

The extent to which the results can be generalised to other situations and people

How can you improve validity in experimental research

Use of a control group, standardised procedures single and double blind trials

How can you improve validity in a questionnaire

Use of a lie scale and anonymity to reduce social desirability

How can you improve validity in observations

Findings maybe more authentic in covert observations