Psychological Assessment

Question 1

What is reliability?

Answer 1

Consistency, reproducibility of an observation (lack of measurement error)

Question 2

What does the Classical Test Theory say?

Answer 2

True Score= Observed Score + Error

Question 3

What are the methods to measure reliability?

Answer 3

Test-retest and equivalent forms

Question 4

What are the methods to estimate reliability?

Answer 4

Split-half, Cronbach’s alpha, Guttman’s lambda, Kuder-Richardson’s KR-20 7 KR-21

Question 5

What form of reliability is computed for qualitative methods such as observations or interviews?

Answer 5

Inter-rater agreement (.005-.25 for observations, .05-0.5 for interviews)

Question 6

What is Spearman-Brown’s Propechy?

Answer 6

True variance increases more rapidly than error variance (reliability usually increases with the number of items)

Question 7

What is the Standard Error of Measurement (SEM) used for?

Answer 7

It is used for estimation of measurement error and for comparison issues.

Question 8

What is the formula for SEM?

Answer 8

SEM= S sqrt(1- alpha)
s= standard deviation

Question 9

The formula for the standard error of prediction is computed the same as for the standard error of measurement, except the alpha is the test-retest correlation. True or false?

Answer 9

True

Question 10

Reliability is a condition for validity. True or false?

Answer 10

True

Question 11

What is validity?

Answer 11

A complex and integrated corpus of demonstrations/scientific knowledge that examines the psychological variables measured by a test.

Question 12

What are the sources of validity?

Answer 12

Content validity (construct, face, translation validity) and criterion-related validity (predictive, concurrent, convergent, divergent/discriminat)

Question 13

What is construct validity?

Answer 13

The extent to which measured dimension are truly relevant to the intended concept => concept underrepresentation or irrelevance? + do the dimension aggregate the way the theory prescribes? (exploratory/confirmatory factor analysis)

Question 14

What is criterion validity?

Answer 14

Indicator that test can be used as a predictor for a certain behavior (criterion).

Question 16

How is concurrent validity measured?

Answer 16

T-tests or cluster analysis, G-type factor analysis: contrasting/distinguishing between groups

Question 17

What is the most prevalent case in which divergent validity is needed?

Answer 17

The case of comorbidity

Question 18

Is face validity a good method to infer criterion-related validity (based on items or the nature of the problem)?

Answer 18

No

Question 19

Number of items determines validity if they are well-constructed. True or false?

Answer 19

True (longer tests are more valid)

Question 20

What are the most valid methods?

Answer 20

Assessment center (multiple measures; .65), behavioral interview (.40-.60), work sample (.54), ability test (.53), integrity tests (.51), personality questionnaires (.39), biodata (.38), references (.23), traditional interviews (unstructured/structures; .05-.19)

Question 21

Incremental validity is computed with stepwise hierarchical treatment. True or false?

Answer 21

True

Question 22

Differential validity is also called:

Answer 22

Adverse impact or test bias

Question 23

There is always a trade-off between internal (stastical conclusion validity) and external validity (population or ecological validity). True or false?

Answer 23

True

Question 24

What is the difference between efficacy and effectiveness?

Answer 24

Effectiveness=effect under real world setting
Efficacy=effect under ideal setting