STATS lec 9- Validity Flashcards
1
Q
Validity
A
- Valid: well grounded in logic or truth
- Gives us a true interpretation
2
Q
Forms of validity 1- Face validity
A
- Face validity
- Does it measure what it says it does
- Do the questions in a stats exam reflect the knowledge students should have learnt
- If the outcome wanted is a reduction in heart attacks, is measuring BP a goodway of assessing this?
- Is the way an answer to a question in a survey is interpreted valid
3
Q
Forms of validity 2- Criterion validity
A
-
Criterion validity: does the measure give results which are in agreement with other measures of the same thing?
- e.g. do theoretical research methods exam scores relate to practical research scores
-
Concurrent: comparison of the new test with an established test
- new intelligence test vs IQ
-
Predictive: does the test predict the outcome of some future attempt
- STATS vs A-level results
4
Q
2 types of criterion validity
A
-
Concurrent: Comparison of new test with established test
- e.g. new intelligence test c.f. IQ
-
Predictive: does the test predict the outcome of some future attempt
- E.g. do your STATS scores predict you A level grades
5
Q
Forms of validity- Construct validity
A
-
Construct validity: How well does the test tap into the underlying construct
- e.g. does a stats exam actually measure status knowledge (or is it testing memory)
- Is a score of 30 on a personality scale a good measure of personality changes in schizophernia
6
Q
Forms of validity- Ecological validity
A
- Ecological validity: does the study reflect naturally occuring behaviour
- e.g. would the performance of students in practical labs be equivalent to the performance of those students in a real setting
- Do students in the dispensary react to a prescription in the same way as they do in “real life”
- Would an animals behaviour in captivity tell you anything about animals in the wild
7
Q
Forms of validity- population validity
A
-
Population validity: is our sample adequate for the claims we make about the population
- e.g. if we measured attitudes to recreational drug use form a sample in this lecture, could we generalise the results to conclude the attitudes of everyone/young people/ students/ Aston students/ Pharmacy staff
- Sampling is important
8
Q
Sampling
A
- VERY important, if you get it wrong you can compromise your population validity
- (And therefore your conclusions may be wrong)
- See lec 8 + 9