Psychometrics Quiz
Questions: 16 · 10 minutes
1. On a hiring questionnaire, applicants consistently underreport common but undesirable behaviors to create a favorable impression. What is the most likely concern?
Social-desirability or impression-management bias
Inter-rater disagreement
Practice effects
Restriction of range
2. What is psychometrics primarily concerned with?
The treatment of psychological disorders through medication
The theory and practice of measuring psychological attributes
The biological classification of brain structures
The study of group behavior without using measurement
3. A measure of social anxiety relates strongly to established social-anxiety measures, less strongly to general stress, and minimally to an unrelated physical skill. What does this pattern mainly support?
Perfect test-retest reliability
Standardized administration across testing sites
Construct validity through expected convergent and discriminant relationships
The absence of all response bias
4. A new questionnaire has 20 items intended to measure one narrow construct. The researcher checks whether the items tend to produce coherent response patterns. What is being assessed?
Face validity
Test-retest reliability
Criterion contamination
Internal consistency
5. In psychometrics, what does validity refer to?
How quickly a test can be administered and scored
How well evidence supports the intended interpretation and use of scores
How similar all the questions appear on the surface
How often the same people receive identical raw scores
6. Two people have observed scores that differ by 2 points on a test whose standard error of measurement is 4 points. What is the most cautious interpretation?
The higher-scoring person certainly has more of the measured attribute
The difference proves the test has no reliability
The difference is small relative to measurement uncertainty and should not be overinterpreted
The scores must be converted to percentiles before any uncertainty exists
7. Why are standardized administration instructions important?
They ensure that every participant earns the same score
They automatically make the norm group representative
They prove that a test is appropriate in every cultural setting
They reduce unwanted score differences caused by inconsistent testing conditions
8. A licensing examination requires candidates to demonstrate a defined level of competence, regardless of how other candidates perform. What kind of interpretation is this?
Norm-referenced interpretation
Ipsative interpretation
Percentile-based interpretation
Criterion-referenced interpretation
9. On a standardized distribution with a mean of 0 and a standard deviation of 1, what does a z-score of +1.5 indicate?
A score 1.5 standard deviations above the mean
A score 1.5 raw points above the highest possible score
A score achieved by exactly 1.5% of the reference group
A score with 1.5 times the test's reliability
10. Two trained observers independently code the same interviews. Their ratings are then compared. Which type of reliability is most relevant?
Inter-rater reliability
Parallel-forms reliability
Test-retest reliability
Split-half reliability
11. A selection test is administered to applicants, and their scores are later compared with job-performance ratings. A meaningful association would provide which kind of evidence?
Face-validity evidence only
Internal-consistency evidence
Content-validity evidence only
Predictive criterion-related validity evidence
12. A bathroom scale gives the same reading each morning but is always two kilograms too high. Which description fits best?
The scale has high validity because its error is consistent
The scale is valid but unreliable
The scale is reliable but lacks accuracy as evidence of validity
The scale cannot be evaluated because reliability and validity are unrelated
13. A researcher gives the same stable-trait questionnaire to the same group twice, three weeks apart. Which property is being examined most directly?
Content validity
Test-retest reliability
Inter-rater reliability
Predictive validity
14. People from two demographic groups have the same overall ability level, but one test item is systematically harder for one group. What should researchers investigate?
Possible differential item functioning and item bias
Whether more difficult items always have greater validity
Whether the higher-scoring group should define the passing standard
Whether all group differences can be removed by changing raw scores to percentiles
15. Which statement best describes a reliable test?
It measures every relevant aspect of a broad construct
It produces reasonably consistent scores under comparable conditions
It guarantees an unbiased decision for every test taker
It predicts every important real-world outcome accurately
16. A test report places a student at the 75th percentile in the reference group. What does this mean?
The student's score was 75 points above the group average
The student answered exactly 75% of the questions correctly
The student scored as high as or higher than about 75% of that reference group
The student has mastered 75% of the measured skill