reliability Flashcards
(24 cards)
What is reliability?
A measure of consistency. If a measurement is made twice and produces the same result, it’s considered reliable.
Give an everyday example of reliability.
A ruler should find the same measurement for an object every time (unless broken).
What is the test-retest method?
Administering the same test or questionnaire to the same person on different occasions.
What indicates good reliability in the test-retest method?
The results obtained should be the same or very similar each time.
When is the test-retest method commonly used?
With questionnaires and psychological tests (e.g. IQ tests).
What must be considered when using test-retest?
Sufficient time between tests so participants don’t recall answers but not so long that attitudes change.
What does a significant positive correlation show in test-retest?
The reliability of the measuring instrument is good.
What is inter-observer reliability?
Ensuring different observers interpret events in the same way.
Why is inter-observer reliability important?
To avoid subjectivity, bias and unreliability.
How can inter-observer reliability be established?
Conducting observations in teams of at least two and correlating the data.
What is a pilot study in the context of observations?
A small-scale trial run to check that observers apply behavioral categories consistently.
Give an example of when we would use inter-rater reliability.
Similar methods applied to other forms of observation
How is reliability measured?
Using a correlational analysis.
What correlation coefficient is desired for reliability?
The correlation coefficient should exceed +.80.
How can the reliability of questionnaires be improved?
By using the test-retest method and ensuring a correlation that exceeds +.80.
What should be done if a questionnaire has low test-retest reliability?
Some items may need to be ‘deselected’ or rewritten if they are complex or ambiguous.
How can the reliability of interviews be improved?
Use the same interviewer each time
What type of interviews are more reliable?
Structured interviews where the interviewer’s behavior is more controlled.
How can the reliability of observations be improved?
By ensuring behavioral categories have been properly operationalized and are measurable.
What should be avoided when creating behavioral categories?
Categories that overlap.
What happens if categories are not operationalized well?
Observers may make differing and inconsistent judgments.
What should observers do if reliability is low?
Receive further training or discuss their decisions to apply categories consistently.
What is the focus of reliability in an experiment?
The procedures.
Why are standardized procedures important in experiments?
To compare the performance of different participants and studies.
