SIOP Psychological Assessment & Measurement Techniques 4 — Questions and Answers
Question 1: The Standard Error of Measurement (SEM) is used primarily to:
- Estimate the population mean from a sample
- Construct confidence intervals around an individual's observed score (Correct answer)
- Correct validity coefficients for range restriction
- Calculate the probability of adverse impact
Correct answer: Construct confidence intervals around an individual's observed score
The SEM quantifies measurement error and is used to create confidence intervals around an observed score, expressing the range within which the true score likely falls.
Question 2: Which type of criterion contamination occurs when raters know the predictor score of the employee when evaluating job performance?
- Halo error
- Leniency bias
- Criterion contamination by knowledge of predictor standing (Correct answer)
- Criterion deficiency
Correct answer: Criterion contamination by knowledge of predictor standing
Criterion contamination by predictor knowledge occurs when performance raters are aware of an employee's test scores, which can bias their ratings and inflate predictor-criterion correlations.
Question 3: A biodata instrument validated using the empirical keying approach is MOST susceptible to which threat?
- Construct under-representation
- Capitalization on chance (sample-specific validity) (Correct answer)
- Social desirability bias
- Adverse impact against age groups
Correct answer: Capitalization on chance (sample-specific validity)
Empirical keying selects items that correlate with criteria in the development sample, risking capitalization on chance correlations that may not replicate in new samples.
Question 4: In the context of assessment feedback, which ethical principle requires practitioners to disclose assessment limitations and the meaning of scores to test-takers?
- Beneficence
- Nonmaleficence
- Informed consent and transparency (Correct answer)
- Justice
Correct answer: Informed consent and transparency
Informed consent and transparency require that test-takers receive adequate information about how their scores will be used, interpreted, and the limitations of the assessment.
Question 5: Multitrait-multimethod (MTMM) matrices are used to evaluate which aspects of construct validity?
- Convergent and discriminant validity simultaneously (Correct answer)
- Predictive and concurrent validity
- Content and face validity
- Criterion-related and incremental validity
Correct answer: Convergent and discriminant validity simultaneously
MTMM matrices assess both convergent validity (same trait, different methods should correlate highly) and discriminant validity (different traits should not correlate highly regardless of method).
Question 6: When a cognitive ability test is used as a hurdle in a multiple-hurdle selection system, which statistical consequence should the I-O psychologist be aware of?
- The test's validity will increase for those who pass the hurdle
- Range restriction on the cognitive test will attenuate its validity estimate in the selected group (Correct answer)
- Adverse impact is eliminated for all subsequent predictors
- The reliability of the test increases in the restricted sample
Correct answer: Range restriction on the cognitive test will attenuate its validity estimate in the selected group
Using a test as a preliminary hurdle restricts the range of scores in those who advance, attenuating the observed validity coefficient for the cognitive test in that selected sample.
Question 7: Which administration format is MOST susceptible to impression management and deliberate faking on personality assessments?
- Low-stakes developmental feedback context
- High-stakes employment selection context (Correct answer)
- Peer-rating context
- Clinical diagnosis context
Correct answer: High-stakes employment selection context
High-stakes employment selection motivates applicants to present themselves favorably, making deliberate faking and impression management most prevalent in that context.
SIOP Psychological Assessment & Measurement Techniques 4
The Standard Error of Measurement (SEM) is used primarily to: