Reliability/validity Flashcards
Reliability
Links to consistency
Validity
Links to accuracy
Internal reliability
Relates to the extent to which the procedure is standardised and replicable
External reliability
Relates to the extent to which the sample is large enough to establish a consistent effect
Standardised
When a procedure has implemented many controls that means it is the same for every participant
Replicable
When the procedure can be repeated exactly the same again and again
Split-half reliability
When you compare items within a test to see if they are giving the same findings
Inter-rater reliability
When you check one observer’s findings against those of another observer to see if they are consistent with each other
Test-retest
When you repeat they study with the same/very similar participants to see if the same results are obtained the second time
Construct validity (internal)
Whether the test has been shown to measure what is supposed to be measuring
Ecological validity (external)
Whether the study represents a real life situation
Population validity (external)
-Whether the sample used can be generalised to the wider population
-How diverse a sample is
What reduces construct validity?
-Lack of controls
-Demand characteristics
-Participant variables
-Social desirability bias
-Environmental variables
-Order effects
-Obtaining subjective (qualitative) data
What increases ecological validity?
-Conducting the research in a real life (field) setting
Having participants complete a task that is part of their normal life
-If an interview, having an informal, unstructured conversation rather than set questions
Face validity (internal)
Whether, on the face of it, the test looks like it is measuring what is supposed to measure
