RELIABILITY AND VALIDITY

Question 1

what three criteria need to be considered for the assessment of individuals and measures used to research purposes?

Answer 1

1. objectivity - test or measure should yield similar outcomes irrespective of who administers to measures
2. reliability - consistency over different times/circumstances
3. validity - extent to which a measures assesses what it claims to measures e.g. face validity, content, concurrent, construct

Question 2

why can there be measuring instruments that are adequate for research purposes but not for assessing individuals?

Answer 2

1. research based on sample of individuals rather than a singular case, so psych test that discrim between people useful for research but x apply to everyone
2. researcher may be left with new measure/ poorly documented measure as it is available, due to reliability and validity of measure taking long period of time
3. one shouldn’t assume a satisfactory measure for indiviuals as can be applied in diff setting e.g. depression clinical scale for clinical/non-clinical samples
4. less good measure better choice is researcher dealing with large sample as individual assessment takes a long time to administer

Question 3

main sources of info for psych tests are…

Answer 3

1. instruction manual for test
2. books/journal articles about measure
3. catalogues of published tests
4. internet

Question 4

what is internal reliability?

Answer 4

how consistently all of the items in a scale measure the concept in question

Question 5

traditional way of calculating internal reliability is…

Answer 5

calculating score of half of items on test and correcting scores for remainder of test

Question 6

what are 3 measures of alpha reliability?

Answer 6

1. split half reliability - half of items summed, then second half, pearson correlation for both halves
2. odd-even reliability - two halves for odd and even numbered items, correlation calculated
3. alpha reliability (Cronbach’s alpha) - average of all possible split-half reliability’s, gives best overall picture

Question 7

stability over time measures of reliability is…

Answer 7

test-retest reliability - the extent to which measure remains stable between two different points in time

Question 8

equivalent measures reliability is…

Answer 8

alternate forms reliability - extent to which two equivalent versions of the test correlate

Question 9

what should be considered when examining the validity of a test?

Answer 9

1. validity is not property of text but a complex matter of test, sample it’s used on and social context of use
2. there doesn’t need to be a relationship between reliability and validity
3. validity coefficient limited by the reliability of the test

Question 10

what are 9 different types of validity?

Answer 10

face
content
concurrent
known-groups
predictive
construct
triangulation
convergent
discriminant

Question 11

what is face validity?

Answer 11

from the appearance of items, does the scale measure what it claims to?

Question 12

what is content validity?

Answer 12

do items of scale cover important characteristics of concept being measured?

Question 13

what is concurrent validity?

Answer 13

does scale correlate well with other measures of same concept being taken at same time

Question 14

what is known-groups validity?

Answer 14

does measure distinguish between the groups expected?

Question 15

what is predictive validity?

Answer 15

does measure accurately predict future behaviours?

Question 16

what is construct validity?

Answer 16

theoretical and conceptual understanding of thing being measured, how well do we understand it?

Question 17

what is triangulation?

Answer 17

multiple types of measures to assess something, more familiar in qualitative research

Question 18

what is convergent validity?

Answer 18

do different sorts of measures of the same concept correlate?

Question 19

what is discriminant validity?

Answer 19

do measures of apparently different concepts not correlate?