Reliability Flashcards
(12 cards)
what is reliability?
Concerned with consistency of the study
Internal Reliability
The extent to which the research procedures/measurements taken/questions asked are carried out consistently for each participant e.g asking questions in same order and using stopwatch to measure time taken to complete a task etc.)
External Reliability
the extent to which a behaviour is a consistent over time (e.g if a questionnaire on work related stress was completed at weekly intervals for 1 month the findings of each should be similar/consistent.)
Common issues of reliability in experiments
- Subjectivity within the IV/DV leading to inconsistent measurements/conditions
- Multiple Researchers potentially leading to inconsistent measurements
- Non-Standardised Instructions leading to inconsistent behaviour from participants.
How could you solve the issues of reliability in experiments?
-Operationalisation of variables
-A structured design
-Inter rater reliability
-Standardisation of instructions
Common Issues of reliability in self reports
- Subjectivity within questions or answers leading to incorrect answers or interpretation of answers
- Multiple Researchers potentially leading to inconsistent measurements
- Unstructured Design within interviews e.g asking participants different questions.
How could you solve the issues of reliability in self-reports
- Standardisation of Instructions
- Structured Design
- Inter-rater reliability
- Coding systems
Common issues of reliability in observations
- Subjectivity within target behaviour potentially leading to inconsistent measurements
- Multiple researchers potentially leading to inconsistent measurements
- Unstructured Designs e.g trying to record all behaviour could lead to gaps in the data.
How could you solve the issues of reliability in observations?
-Operationalisation of variables
-Structured Design
-Coding systems
Inter rater reliability
This involves two or more researchers assessing whether they have produced similar data/results within the study. If the data they have collected/ results are similar it must have been collected using the same consistent approach meaning it is internally reliable.
Split Half Method
This involves comparing a person’s performance on two halves of a questionnaire or test. If a questionnaire is assessing the same thing in all its questions then there should be a close correlation in the scores derived from both halves of the test (a measure of internal reliability.)
test re-test reliability
This measures consistency of responses over two or more occasions. If someone does an IQ test and their score is 100 one week and 80 another week we would know their behaviour is not consistent/ reliable. If scores are the same/similar we can assume the study has external reliability.
