wk 8

Question 1

test retest

Answer 1

testing it one day and seeing if you get the same result on another day

Question 2

inter-rater reliability

Answer 2

one person measuring it then a different person measures it to see if they get the same result
use Cohen’s Kappa

Question 3

inter-method reliability

parallel forms

Answer 3

will you get the same result if you use 2 different tools to measure (e.g measuring height with a ruler vs tape measure)

use Kappa Coefficient -> look at results of different forms
can also compare means/distribution

Question 4

internal consistency reliability

Answer 4

did you respond to similar questions in a similar way

used the split half technique-> test consistency across 2 halves

Cronbach’s alpha - test the consistency across all items

Question 5

construct validity

Answer 5

is this a valid means of tapping into the construct you are attempted to measure

tested via:

• predictive validity (is it able to predict behaviour in the future)
• concurrent validity (is the score related to another criterion of this construct tested at the same time)

Question 6

convergant validity

Answer 6

is the score related to other measures of the same construct

Question 7

discriminant validity

Answer 7

is the score different from measures that theoretically measure something else

Question 8

content validity

Answer 8

is this a systematic means of measuring the variable?

Question 9

face validity

Answer 9

does the item have face value

Question 10

cost benefit

Answer 10

✔️more likely to be valid than non standardised measures
✔️allows for inclusion of an eclectic mix of questions
✔️can contain items that lack face validity making it harder to fake
✖️only as good as the outcome measures
✖️if done poorly there is no basis for assuming validity
✖️it could contain items you shouldn’t include

Question 11

factor analytic approach

Answer 11

gather information -> run a factor analysis (which variables link together) -> name the factors

analyses variance to find the lowest common denominator

what are the underlying components of the construct being measured

Question 12

cost- benefit of factor analysis

Answer 12

✔️ reduces to a manageable set of items
✔️ helps refine existing tests
✔️ can lead to useful insights about behaviour
✖️limited by the original items used
✖️ a human has to decide how they are conceptually related
✖️ sometimes the factors don’t make logical sense

Question 13

predictive validity

Answer 13

the extent to which you can predict behaviour in the future