Reliability & Validity

Question 1

Definition of reliability

Answer 1

consistency, stability, and accuracy of the instrument

Question 2

Types of reliability

Answer 2

Internal reliability, test-retest reliability, inter-examiner reliability

Question 3

Internal reliability

Answer 3

constancy of result on items within a test/on the overall test. Can be measured using split-half reliability, want a correlation of .7 or more.

Question 4

Test-retest reliability

Answer 4

Stability of the measure across time. Examine relationship of scores on a test over an interval which change is unlikely to occur. Can balance learning effects with spontaneous recovery

Question 5

Inter-examiner reliability

Answer 5

How much the examiner influences performance. Examine relationship between the same individual’s scores on tests given by different examiners

Question 6

Definition of validity

Answer 6

degree to which a test measures what it purports to measure

Question 7

Types of validity

Answer 7

Face validity, Content validity, Criterion validity (concurrent, predictive), Construct validity (convergent, divergent)

Question 8

Face validity

Answer 8

subjective judgments by individuals using the test that there is a match between the purpose of the test and its content

Question 9

Content validity

Answer 9

adequacy of sampling from the domain of the construct to be measured (i.e., items should be representative of the construct)

Question 10

Criterion validity

Answer 10

how well the score on the test can predict a certain outcome. Correlation between the scores on the test variable under study and an external measure

Question 11

Concurrent validity

Answer 11

relationship between a specific criteria and scores from the test (i.e., previously validated test, how well the test discriminates)

Question 12

Predictive validity

Answer 12

how well a score on a test can predict future events

Question 13

Sensitivity

Answer 13

percent of individuals scoring below the cut-off point

Question 14

Specificity

Answer 14

percent of individuals without aphasia scoring above the cut-off point

Question 15

Construct validity

Answer 15

tests ability to index the underlying theoretical construct it is intending to measure. Supported by theoretical rationale and empirical evidence

Question 16

Convergent validity

Answer 16

relationship of test to other tests measuring the same construct (should be high)

Question 17

Divergent validity

Answer 17

relationship of tests to other tests measuring different constructs (should be low)