Research Methods - Validty And Reliability Flashcards
What is interval validity
Does a test/ investigation measure what it was supposed to measure?
What is external validity ?
Can the results of an investigation be generalised beyond the research setting within which it was found?
Explain the three types of external validity
Temporal-The extent to which findings from a research study can be generalised to other historical times and eras.
• Ecological-The extent to which findings from a research study can be generalised to other settings and situations.
• Population-The extent to which findings from a research study can be generalised to other people outside the sample.
What is face validity be done
When A measure is scrutinised to determine whether it appears to measure what it is supposed to measure- e.g. does a test of anxiety actually look like it measures anxiety?
• Done by ‘eyeballing’ the measuring instrument or passing it to an expert to check.
How does concurrent validity work
The extent to which a psychological measure relates to an existing similar measure.
• Close agreement between the two sets of data would indicate the new test has high concurrent
validity.
• Close agreement is indicated if the correlation between the two sets of scores exceeds +.8
What is reliability
How consistent the findings from an investigation are. A measuring device is said to be reliable if it produces consistent findings every time it is used.
What is the test retest way of assessing reliability
Administer same test or questionnaire to the same person on different occasions
Same or similar results obtained
How should be taken in account in test , retest method
Sufficient time between test and retest
Participant cannot recall answers
Not too long that attitudes or opinion have changed
What should be done in last step of test and retest method
Both sets of scores should be correlated to make sure they are significant
Explain inter observer reliability
The extent to which there is an agreement between two or more observers involved in the observations of a behaviour. This is measured by correlating the observations of two or more observers. If the correlation between the two observers is +0.8, the data is said to have high inter observer reliability.
How can you test and improve reliability in questionnaires
Test- retest method
Data from both sets of tests should be correlated to produce correlation that exceed +.0.8
Items deselected, questions rewritten
Too complex, interpretation differently by people
Open questions with closed questions
In experiments how would you increase reliability
Conducting lab experiment
Higher control over variables
How do you improve the reliability of observations
Behavioural categories are properly operationalised
Shouldn’t overlap
Eg hugging and cuddling
How do you increase validity of interviews
Same interviewer each time
If not - they must be trained
To prevent one interviewer from asking too leading questions
Structured interviews - controlled behaviour
How to improve validity in experimental research
Use a control group to assess whether changes in DV are of IV
Standardise procedures
Minimise investigator effects
Single blind and double blind - reduce demand characteristics
External- conduct field experiment
