WGU C207
Data-Driven Decision Making
Week 8 Final Exam
Case Studies & Rationales
Original applied decision-making
study resource
WGU C207 Data-Driven Decision Making
Week 8 Final Exam Case Studies &
Rationales
Original practice resource for study use only — not an official WGU examination or answer
key. Verify concepts against current course materials, instructor guidance, and applicable
organizational policies.
A practical, original case-study collection for interpreting evidence, evaluating uncertainty,
protecting data, and making defensible organizational decisions.
How to use this resource Decision lens
Read each case before viewing the A strong answer connects data quality,
options, identify the decision question, statistical reasoning, operational context,
state any assumptions, and explain what stakeholder needs, and ethical use of
additional evidence would change your information.
recommendation.
Study map
Foundations and data quality · Statistics and uncertainty · Relationships and forecasting ·
Dashboards and operational decisions · Ethics and governance · Integrated recommendations
,Original practice resource | Study use only
6. Integrated Evidence-Based Recommendations
Section focus: Combine statistical evidence, operational judgment, stakeholder priorities,
and ethical safeguards into an actionable recommendation.
Case Study 21. A hospital considers a text-message reminder program. A pilot shows a 4-
percentage-point increase in attendance, but the confidence interval is 1 to 7 points. The
program requires translation and accessibility work for several patient groups. What is the
best recommendation?
A. Scale immediately without further review because the interval is entirely positive.
B. Reject the program because the effect is not exactly 4 points in every setting.
C. Conduct a phased implementation with subgroup monitoring, accessibility safeguards,
cost review, and predefined success measures.
D. Report only the highest subgroup result.
Correct answer: C. Conduct a phased implementation with subgroup monitoring,
accessibility safeguards, cost review, and predefined success measures.
Rationale: The interval supports a likely positive effect but still includes meaningful variation.
A phased approach tests feasibility, equity, cost, and real-world performance without
overclaiming. The alternatives either ignore implementation risk or misuse uncertainty.
Case Study 22. A public health team sees that emergency visits are higher in neighborhoods
with fewer primary-care appointments. Before recommending a new clinic, what should the
team do?
A. Treat the association as proof that adding a clinic will reduce emergency visits.
B. Examine access barriers, population differences, time trends, competing explanations, and
evidence from comparable interventions before selecting a strategy.
C. Ignore emergency visits because they are an outcome rather than a process measure.
D. Choose the neighborhood with the highest rate without checking denominator size.
Correct answer: B. Examine access barriers, population differences, time trends, competing
explanations, and evidence from comparable interventions before selecting a strategy.
Rationale: The association is a useful signal but not proof of a causal effect. Denominators,
confounding, service capacity, patient preferences, and comparable evidence should inform
,the recommendation.
Case Study 23. A software vendor offers a predictive model that flags employees who may
leave. The model is accurate overall but substantially less accurate for a small demographic
group. What should the organization do before deployment?
A. Deploy it because overall accuracy is sufficient.
B. Evaluate subgroup performance, fairness, explainability, privacy, human oversight, and
potential employment harms; revise or decline deployment if risks are unacceptable.
C. Hide subgroup results to prevent confusion.
D. Let the model make termination decisions automatically.
Correct answer: B. Evaluate subgroup performance, fairness, explainability, privacy, human
oversight, and potential employment harms; revise or decline deployment if risks are
unacceptable.
Rationale: Overall accuracy can conceal disparate error rates and unfair impact. Responsible
analytics requires subgroup testing, governance, proportional use, transparency, appeal
paths, and accountable human decisions.
Case Study 24. A leadership team asks for a recommendation after a dashboard shows
worsening turnover, lower engagement, and rising overtime. Interviews suggest workload and
scheduling instability, but the data are incomplete. What is the strongest next step?
A. Recommend a permanent staffing increase immediately based on the dashboard alone.
B. Combine the quantitative trends with validated qualitative findings, check data
completeness, test likely drivers, and propose a monitored pilot with decision thresholds.
C. Ignore interviews because only numerical data count.
D. Wait until the dashboard is perfect before taking any action.
Correct answer: B. Combine the quantitative trends with validated qualitative findings,
check data completeness, test likely drivers, and propose a monitored pilot with decision
thresholds.
Rationale: Triangulation combines complementary evidence while acknowledging
uncertainty. A monitored pilot permits action without pretending the cause is proven, and
explicit thresholds define when to continue, adapt, or stop.
, Concise Answer Key
1–4: 1 B, 2 C, 3 A, 4 C
5–8: 5 C, 6 B, 7 A, 8 B
9–12: 9 B, 10 B, 11 B, 12 B
13–16: 13 B, 14 B, 15 A, 16 B
17–20: 17 B, 18 A, 19 B, 20 B
21–24: 21 C, 22 B, 23 B, 24 B
Study-use note
This resource is original and educational. It is not official WGU examination content, an official
answer key, leaked material, or institution-endorsed guidance. Use it to practice reasoning,
then verify concepts against current C207 course materials, instructor guidance, evidence-
based references, and applicable organizational policies.
Data and ethics note
Protect confidentiality, use approved access and storage controls, document definitions and
assumptions, check data quality and missingness, examine subgroup effects, and avoid using
a single metric without context. Statistical evidence should support—not replace—
professional judgment, stakeholder engagement, and accountable decision-making.
Original practice resource | Study use only
5. Data Ethics, Privacy, Governance, and Communication
Section focus: Protect confidentiality, document assumptions, govern access, and
communicate evidence without overstating certainty.
Case Study 17. A manager asks an analyst to send a spreadsheet containing patient-level
appointment data to a personal email account so the manager can work from home. What is
the best response?
A. Send the file because the manager has a legitimate business role.
B. Use approved secure systems, confirm minimum necessary access, and de-identify or
aggregate data when individual detail is not needed.
C. Add a password to the spreadsheet and use personal email anyway.
Data-Driven Decision Making
Week 8 Final Exam
Case Studies & Rationales
Original applied decision-making
study resource
WGU C207 Data-Driven Decision Making
Week 8 Final Exam Case Studies &
Rationales
Original practice resource for study use only — not an official WGU examination or answer
key. Verify concepts against current course materials, instructor guidance, and applicable
organizational policies.
A practical, original case-study collection for interpreting evidence, evaluating uncertainty,
protecting data, and making defensible organizational decisions.
How to use this resource Decision lens
Read each case before viewing the A strong answer connects data quality,
options, identify the decision question, statistical reasoning, operational context,
state any assumptions, and explain what stakeholder needs, and ethical use of
additional evidence would change your information.
recommendation.
Study map
Foundations and data quality · Statistics and uncertainty · Relationships and forecasting ·
Dashboards and operational decisions · Ethics and governance · Integrated recommendations
,Original practice resource | Study use only
6. Integrated Evidence-Based Recommendations
Section focus: Combine statistical evidence, operational judgment, stakeholder priorities,
and ethical safeguards into an actionable recommendation.
Case Study 21. A hospital considers a text-message reminder program. A pilot shows a 4-
percentage-point increase in attendance, but the confidence interval is 1 to 7 points. The
program requires translation and accessibility work for several patient groups. What is the
best recommendation?
A. Scale immediately without further review because the interval is entirely positive.
B. Reject the program because the effect is not exactly 4 points in every setting.
C. Conduct a phased implementation with subgroup monitoring, accessibility safeguards,
cost review, and predefined success measures.
D. Report only the highest subgroup result.
Correct answer: C. Conduct a phased implementation with subgroup monitoring,
accessibility safeguards, cost review, and predefined success measures.
Rationale: The interval supports a likely positive effect but still includes meaningful variation.
A phased approach tests feasibility, equity, cost, and real-world performance without
overclaiming. The alternatives either ignore implementation risk or misuse uncertainty.
Case Study 22. A public health team sees that emergency visits are higher in neighborhoods
with fewer primary-care appointments. Before recommending a new clinic, what should the
team do?
A. Treat the association as proof that adding a clinic will reduce emergency visits.
B. Examine access barriers, population differences, time trends, competing explanations, and
evidence from comparable interventions before selecting a strategy.
C. Ignore emergency visits because they are an outcome rather than a process measure.
D. Choose the neighborhood with the highest rate without checking denominator size.
Correct answer: B. Examine access barriers, population differences, time trends, competing
explanations, and evidence from comparable interventions before selecting a strategy.
Rationale: The association is a useful signal but not proof of a causal effect. Denominators,
confounding, service capacity, patient preferences, and comparable evidence should inform
,the recommendation.
Case Study 23. A software vendor offers a predictive model that flags employees who may
leave. The model is accurate overall but substantially less accurate for a small demographic
group. What should the organization do before deployment?
A. Deploy it because overall accuracy is sufficient.
B. Evaluate subgroup performance, fairness, explainability, privacy, human oversight, and
potential employment harms; revise or decline deployment if risks are unacceptable.
C. Hide subgroup results to prevent confusion.
D. Let the model make termination decisions automatically.
Correct answer: B. Evaluate subgroup performance, fairness, explainability, privacy, human
oversight, and potential employment harms; revise or decline deployment if risks are
unacceptable.
Rationale: Overall accuracy can conceal disparate error rates and unfair impact. Responsible
analytics requires subgroup testing, governance, proportional use, transparency, appeal
paths, and accountable human decisions.
Case Study 24. A leadership team asks for a recommendation after a dashboard shows
worsening turnover, lower engagement, and rising overtime. Interviews suggest workload and
scheduling instability, but the data are incomplete. What is the strongest next step?
A. Recommend a permanent staffing increase immediately based on the dashboard alone.
B. Combine the quantitative trends with validated qualitative findings, check data
completeness, test likely drivers, and propose a monitored pilot with decision thresholds.
C. Ignore interviews because only numerical data count.
D. Wait until the dashboard is perfect before taking any action.
Correct answer: B. Combine the quantitative trends with validated qualitative findings,
check data completeness, test likely drivers, and propose a monitored pilot with decision
thresholds.
Rationale: Triangulation combines complementary evidence while acknowledging
uncertainty. A monitored pilot permits action without pretending the cause is proven, and
explicit thresholds define when to continue, adapt, or stop.
, Concise Answer Key
1–4: 1 B, 2 C, 3 A, 4 C
5–8: 5 C, 6 B, 7 A, 8 B
9–12: 9 B, 10 B, 11 B, 12 B
13–16: 13 B, 14 B, 15 A, 16 B
17–20: 17 B, 18 A, 19 B, 20 B
21–24: 21 C, 22 B, 23 B, 24 B
Study-use note
This resource is original and educational. It is not official WGU examination content, an official
answer key, leaked material, or institution-endorsed guidance. Use it to practice reasoning,
then verify concepts against current C207 course materials, instructor guidance, evidence-
based references, and applicable organizational policies.
Data and ethics note
Protect confidentiality, use approved access and storage controls, document definitions and
assumptions, check data quality and missingness, examine subgroup effects, and avoid using
a single metric without context. Statistical evidence should support—not replace—
professional judgment, stakeholder engagement, and accountable decision-making.
Original practice resource | Study use only
5. Data Ethics, Privacy, Governance, and Communication
Section focus: Protect confidentiality, document assumptions, govern access, and
communicate evidence without overstating certainty.
Case Study 17. A manager asks an analyst to send a spreadsheet containing patient-level
appointment data to a personal email account so the manager can work from home. What is
the best response?
A. Send the file because the manager has a legitimate business role.
B. Use approved secure systems, confirm minimum necessary access, and de-identify or
aggregate data when individual detail is not needed.
C. Add a password to the spreadsheet and use personal email anyway.
