Evaluating Research(research methods) Flashcards
What is vailidity
Does the study really measure what it’s supposed to measure ?
Internal validity
Are ppts behaving in a natural way ?
Ecological validity
To what extent does the measure truly reflect behaviour in real life settings
- how far is the environment similar to what a person does in real life
- how far is the task similar to what a person does in real life
Face validity
Does the measure actually seem to be measuring what it’s says it’s measuring
Population validity
Refers to whether the sample is representative of the wider target population of the study
Construct validity
The extent to which an instrument used for making a diagnoses actually includes a full range of items which match well with those listed in the DSM or ICD
Concurrent validity
The extent to which the result of study or test is the same as the results of other studies or tests done at the same time, but using an alternative way of reaching a diagnosis
Criterion validity
The extent to which a measure can predict the performance of behaviour of the measured thing
Factors influencing reliability
- high levels of control
- standardised instructions
- standardised materials
- standardised procedures
External reliability
If a test produces the same results each time the test is carried out. This can be oooked at using test- retest method
Internal reliability
The extent to which a measure is consistent within itself, such as psychometric tests and questionnaires. This can be assessed unsung the split half method
This involves splitting a test into two and gavin. The same ppt doing both halves of the test. If the two halves of the test provide similar results this would suggest that the test has internal validity
Inter- rater reliability
When different researchers interpret results consistently/ agree
Improving reliability
When possible take more than one measurement from each ppt
Pilot studies
Standardise the way in which researchers collect data, inter-rater reliability
Check data very carefully
