Validity And Reiablity Flashcards
What is the definition of reliability?
Measure of consistency.
If the study is repeated it will produce the same results.
What are the ways of assessing reliability?
- Test-retest
- Inter-observer reliability
What is Test-retest?
- Giving the same test/ questionnaire to the same person on different occasions.
What must there be between test-retests?
Time so that answers aren’t recalled.
What happens to the scores of test-retest?
Sets of scores are correlated - if the correlations is significant and positive then reliability is assumed to be good.
What should happen during inter-observer reliability?
- Should conduct observers with not her researcher.
- Behavioural categories should be applied the same way.
- Can be correlated
How can we improve the reliability of questionnaires?
- Test-retest.
- Comparing sets of data should produce a correlation that exceeds +80.
- Some questions may have to be re-written if they have low test-retest reliability.
- Some questions may be hard to interpret.
How do we improve the reliability of interviews?
- The same interviewer each time.
- Questions aren’t too leading or ambitious.
How do we improve the reliability of observations?
- Behavioural categories should be properly operationalised.
- Categories should be less open to interpretation and not overlap.
Improving experiments reliability.
- Lab experiments have the highest control and show precise replication rather than reliability of a finding.
What does validity mean?
Whether a test produces a result that’s legitimate.
What is internal validity?
Whether effects in an experiment are due to manipulation of IV or another factor.
What is external validity?
Factors outside investigation and whether the results can be applied to the population/ settings.
Ecological validity
Generalising findings from one setting to another - usually real life
What is temporal validity?
Whether finds hold true over time