, Question 1
A D396 Task 2 submission includes a 10-item assessment where all items are
traditional four-option multiple-choice. According to NCSBN's Clinical
Judgment Measurement Model and Next Gen NCLEX item taxonomy, which
revision most directly increases construct coverage of 'Evaluate Outcomes'?
A. Convert two recall items into matrix/grid items requiring ranking of
nursing actions.
B. Add two unfolding case-study items with embedded bowtie and trend
exhibits.
C. Replace three items with extended multiple-response
(select-all-that-apply) formats.
D. Increase the number of options per item from four to five to reduce
guessing.
Correct Answer: B - Add two unfolding case-study items with
embedded bowtie and trend exhibits.
RATIONALE
Evaluate Outcomes is best measured through unfolding case studies
that require interpreting evolving clinical data (trends, bowtie) - a
hallmark of Next Gen NCLEX. Matrix and SATA items primarily
target Recognize Cues/Analyze Cues, and adding options does not
change the cognitive construct.
Question 2
A D396 Task 2 rubric awards points for 'alignment of assessment items to
course objectives.' Which psychometric property is most directly threatened
when items are aligned to objectives but all items share a single cognitive level
(e.g., recall)?
A. Test-retest reliability
B. Construct validity
C. Interrater reliability
Page 2
, D. Internal consistency (KR-20)
Correct Answer: B - Construct validity
RATIONALE
Construct validity is threatened when the item pool underrepresents
the intended cognitive domain (e.g., only recall when objectives span
application/analysis). Reliability indices like KR-20 or test-retest can
remain high even when the construct is narrowly sampled.
Question 3
In a D396 Task 2 submission, a candidate proposes using a 5-point Likert scale
to measure 'clinical judgment' in a simulation. Which critique is most
defensible under AERA/APA/NCME Standards?
A. Likert scales are inherently invalid for any nursing construct.
B. A single global Likert rating cannot capture the multi-dimensional,
process-based nature of clinical judgment without a validated rubric.
C. Likert scales should only be used with at least 10 points to ensure
interval-level data.
D. Clinical judgment must be measured with physiological biomarkers,
not self-report.
Correct Answer: B - A single global Likert rating cannot capture
the multi-dimensional, process-based nature of clinical judgment
without a validated rubric.
RATIONALE
Clinical judgment is a multi-dimensional, process-oriented construct
(Tanner's model; NCSBN CJMM) that requires validated behavioral
rubrics, not a single global Likert item. The other options overstate or
misapply measurement principles.
Page 3
, Question 4
A candidate's D396 Task 2 includes a table of test blueprints. The table shows
60% of items on 'Safety' and 5% on 'Patient Education,' while course objectives
weight these domains equally. Which action best demonstrates synthesis of
blueprint principles?
A. Increase total item count to reduce sampling error.
B. Re-weight items to match the objective-to-item ratio and document the
rationale.
C. Remove the Patient Education domain since it is underrepresented.
D. Add more Safety items to strengthen the exam's reliability.
Correct Answer: B - Re-weight items to match the
objective-to-item ratio and document the rationale.
RATIONALE
Blueprint alignment requires item distribution to reflect the relative
emphasis of course objectives. Re-weighting with documented
rationale restores content validity; simply adding items or deleting a
domain does not correct the misalignment.
Question 5
A candidate's D396 Task 2 uses a case study where the client's condition
changes across three time points. Which item type most rigorously assesses
'Generate Solutions' as defined by the NCSBN CJMM?
A. A single best-answer question asking for the priority nursing
diagnosis.
B. A matrix item where the candidate selects all appropriate interventions
from a list.
C. A bowtie item requiring the candidate to link actions to parameters to
evaluate.
D. A multiple-choice item asking for the mechanism of action of a
prescribed drug.
Page 4
A D396 Task 2 submission includes a 10-item assessment where all items are
traditional four-option multiple-choice. According to NCSBN's Clinical
Judgment Measurement Model and Next Gen NCLEX item taxonomy, which
revision most directly increases construct coverage of 'Evaluate Outcomes'?
A. Convert two recall items into matrix/grid items requiring ranking of
nursing actions.
B. Add two unfolding case-study items with embedded bowtie and trend
exhibits.
C. Replace three items with extended multiple-response
(select-all-that-apply) formats.
D. Increase the number of options per item from four to five to reduce
guessing.
Correct Answer: B - Add two unfolding case-study items with
embedded bowtie and trend exhibits.
RATIONALE
Evaluate Outcomes is best measured through unfolding case studies
that require interpreting evolving clinical data (trends, bowtie) - a
hallmark of Next Gen NCLEX. Matrix and SATA items primarily
target Recognize Cues/Analyze Cues, and adding options does not
change the cognitive construct.
Question 2
A D396 Task 2 rubric awards points for 'alignment of assessment items to
course objectives.' Which psychometric property is most directly threatened
when items are aligned to objectives but all items share a single cognitive level
(e.g., recall)?
A. Test-retest reliability
B. Construct validity
C. Interrater reliability
Page 2
, D. Internal consistency (KR-20)
Correct Answer: B - Construct validity
RATIONALE
Construct validity is threatened when the item pool underrepresents
the intended cognitive domain (e.g., only recall when objectives span
application/analysis). Reliability indices like KR-20 or test-retest can
remain high even when the construct is narrowly sampled.
Question 3
In a D396 Task 2 submission, a candidate proposes using a 5-point Likert scale
to measure 'clinical judgment' in a simulation. Which critique is most
defensible under AERA/APA/NCME Standards?
A. Likert scales are inherently invalid for any nursing construct.
B. A single global Likert rating cannot capture the multi-dimensional,
process-based nature of clinical judgment without a validated rubric.
C. Likert scales should only be used with at least 10 points to ensure
interval-level data.
D. Clinical judgment must be measured with physiological biomarkers,
not self-report.
Correct Answer: B - A single global Likert rating cannot capture
the multi-dimensional, process-based nature of clinical judgment
without a validated rubric.
RATIONALE
Clinical judgment is a multi-dimensional, process-oriented construct
(Tanner's model; NCSBN CJMM) that requires validated behavioral
rubrics, not a single global Likert item. The other options overstate or
misapply measurement principles.
Page 3
, Question 4
A candidate's D396 Task 2 includes a table of test blueprints. The table shows
60% of items on 'Safety' and 5% on 'Patient Education,' while course objectives
weight these domains equally. Which action best demonstrates synthesis of
blueprint principles?
A. Increase total item count to reduce sampling error.
B. Re-weight items to match the objective-to-item ratio and document the
rationale.
C. Remove the Patient Education domain since it is underrepresented.
D. Add more Safety items to strengthen the exam's reliability.
Correct Answer: B - Re-weight items to match the
objective-to-item ratio and document the rationale.
RATIONALE
Blueprint alignment requires item distribution to reflect the relative
emphasis of course objectives. Re-weighting with documented
rationale restores content validity; simply adding items or deleting a
domain does not correct the misalignment.
Question 5
A candidate's D396 Task 2 uses a case study where the client's condition
changes across three time points. Which item type most rigorously assesses
'Generate Solutions' as defined by the NCSBN CJMM?
A. A single best-answer question asking for the priority nursing
diagnosis.
B. A matrix item where the candidate selects all appropriate interventions
from a list.
C. A bowtie item requiring the candidate to link actions to parameters to
evaluate.
D. A multiple-choice item asking for the mechanism of action of a
prescribed drug.
Page 4
