Reliability Flashcards
What is reliability
Refers to the consistency or replicability of research.
What are the 5 types of reliability
Internal reliability
External reliability
Inter-rater reliability
Split half method
Test-retest reliability
What is internal reliability
Whether a procedure of a study is standardized so that each participant experiences the same thing
each participant should be tested in the same way
What does internal reliability allow us to check for
We can replicate so it allows us to check for consistency
What is external reliability
the extent to which the results of a procedure can be replicated from one time to another to try and support the findings of a study
What is an example of external reliability
Produces consistent results if repeated again and again. For example if you took an IQ test twice and got 105 for both then its reliable
What is inter-rater reliability
If theres a high inter rater reliability that means 2 or more individuals have a high agreement on a score and therefore the measure of behaviour is reliable (80% of agreement)
What does inter-rater reliability mean for observations
If more than one person observing the same behaviour or different observers watching different individuals should agree on the behaviour measured
How do you establish if a measure has inter-rater reliability
researcher needs to initially compare the results of each researcher to check if they matched. The results from each researcher are then compared using a correlation. If the observers are agreeing and a positive correlation of 80% is established then inter observer reliability is said to be achieved
How do you increase internal reliability
Use a standardised procedure (same always all) there may be some instances when its made difficult to replicate research
Use multiple measures of behavior as these should back each other up and be consistent
How do you increase external reliability
Use measures which generate quantitative data as itll be easier to check for consistency by comparing findings as its unlikely to be misinterpreted when being analysed- no inconsistency
How do you increase external reliability (2)
Different interpretations on a subjective view of rating scales can decrease consistency. Important to check whether ppts understand questions and that its fully operationalised
Use a large sample size so results are less vulnerable to anomalies so more likely to establish consistent trends and patterns
How do you increase inter-rater reliability
Observe or measure behaviour in exactly the same way in the same location for the same amount of time
Run a pilot test first so that researchers are able to measure behaviour in same way
Train researchers on using measuring tools ensure they understand exactly how to use them and what they mean
What is the split half method
Applies mainly to self reports so results in 2 halves of the questionnaire are similar so we can assume test is reliable. For exmaple putting repeat questions in a quesstionnaire which may be asked slightly differently to check how reliable people r being
What is the test-retest method
Testing participants more than once. One person should get the same or very similar scores on the same test
May be that the measure is administered to the same group of people twice
