WGU C784 - APPLIED HEALTHCARE STATISTICS
PREASSESSMENT EXAM LATEST VERSION ACTUAL
EXAM QUESTIONS AND DETAILED ANSWERS - 180
Questions and Answers Already Graded A+ Premium Exam
Tested And Verified
Subject Area Applied Healthcare Statistics
Description This preassessment exam evaluates mastery of statistical methods applied to
healthcare data, including probability distributions, hypothesis testing, regression
analysis, and experimental design. It emphasizes interpretation of results and
application to clinical decision-making.
Expected Grade A+
Total Questions 180
Duration 3 hours
Learning Outcomes 1. Apply descriptive and inferential statistics to healthcare data
2. Interpret statistical results in clinical contexts
3. Design and analyze studies using appropriate statistical tests
4. Evaluate evidence-based practice using statistical reasoning
Accreditation This exam meets the rigor standards of top US R1 universities (e.g., Harvard,
Stanford) and accredited healthcare programs.
Page 1
,1. A healthcare researcher is analyzing the association between a new biomarker
(continuous) and disease progression (ordinal: mild, moderate, severe). Which
statistical method is most appropriate for assessing this association while adjusting
for confounding variables?
A. Pearson correlation coefficient
B. Spearman rank correlation
C. Ordinal logistic regression
D. Multinomial logistic regression
Answer: C. Ordinal logistic regression
Ordinal logistic regression (proportional odds model) is designed for an ordinal
outcome and can adjust for confounders. Spearman correlation only assesses bivariate
monotonic association without adjustment. Pearson requires continuous normal data.
Multinomial logistic regression treats outcome as nominal, ignoring ordering.
2. In a randomized controlled trial comparing two treatments, the primary outcome
is binary (success/failure). A logistic regression model yields an odds ratio of 1.5 for
treatment A vs B (p=0.03). Which interpretation is correct?
A. The relative risk of success is 1.5 for treatment A compared to B.
B. The odds of success are 1.5 times higher for treatment A than B, but this does not directly
equal relative risk.
C. Treatment A reduces the odds of failure by 50% relative to B.
D. The probability of success is 1.5 times greater for treatment A.
Answer: B. The odds of success are 1.5 times higher for treatment A than B, but
this does not directly equal relative risk.
Odds ratio approximates relative risk only when the outcome is rare. Here, without
incidence information, OR cannot be interpreted as risk ratio. Option B correctly
distinguishes odds from probability. Option D mistakes odds for probability. Option C
misstates direction (1.5 OR means higher odds of success, not lower failure).
Page 2
,3. A study reports a 95% confidence interval for the mean reduction in blood
pressure after a new drug: (-2.3, 5.1) mmHg. Which conclusion is most appropriate?
A. The drug significantly reduces blood pressure because the interval includes zero.
B. The drug does not have a statistically significant effect at =0.05 because the interval
includes zero.
C. The drug reduces blood pressure by an average of 2.3 to 5.1 mmHg with 95% confidence.
D. There is a 95% probability that the true mean reduction is between -2.3 and 5.1 mmHg.
Answer: B. The drug does not have a statistically significant effect at =0.05 because
the interval includes zero.
A confidence interval that includes zero indicates no statistically significant difference
at the corresponding level. Option A misinterprets inclusion of zero as significance.
Option C misstates the interval bounds. Option D is a common misinterpretation; the
correct interpretation is that 95% of such intervals contain the true mean.
4. In a survival analysis, the Kaplan-Meier estimator is used to compare survival
curves between two groups. The log-rank test yields a p-value of 0.04. Which
assumption is critical for the validity of this test?
A. The hazard ratio is constant over time.
B. Censoring is independent of the event time.
C. Survival times follow an exponential distribution.
D. The two groups have equal sample sizes.
Answer: B. Censoring is independent of the event time.
The log-rank test requires non-informative censoring (censoring independent of event
time). Option A is an assumption for Cox proportional hazards but not log-rank.
Option C is not required; KM is nonparametric. Option D is not necessary; unequal
sample sizes are allowed.
Page 3
, 5. A researcher plans a study to compare the means of three independent groups
using ANOVA. Preliminary data show that the variances are unequal (Levene's test
p=0.01) and the distributions are moderately skewed. Which approach is most
appropriate?
A. Proceed with standard one-way ANOVA because it is robust to violations.
B. Use Welch's ANOVA with a robust post-hoc test (e.g., Games-Howell).
C. Transform the data to achieve normality and equal variances, then use ANOVA.
D. Use a nonparametric Kruskal-Wallis test without further adjustment.
Answer: B. Use Welch's ANOVA with a robust post-hoc test (e.g., Games-Howell).
Welch's ANOVA does not assume equal variances and is more robust than standard
ANOVA when variances are unequal. Games-Howell post-hoc test also does not assume
equal variances. Transformations may not resolve both issues. Kruskal-Wallis is an
option but less powerful and does not directly compare means.
6. A diagnostic test has sensitivity of 95% and specificity of 90%. The prevalence of
the disease in the tested population is 2%. What is the positive predictive value
(PPV) of this test?
A. 0.162
B. 0.172
C. 0.190
D. 0.209
Answer: A. 0.162
PPV = (sensitivity × prevalence) / (sensitivity × prevalence +
(1-specificity)×(1-prevalence)). Plugging: (0.95×0.02)/(0.95×0.02 + 0.10×0.98) =
0.019/(0.019+0.098)=0.019/0.1170.1624. Thus 0.162.
Page 4
PREASSESSMENT EXAM LATEST VERSION ACTUAL
EXAM QUESTIONS AND DETAILED ANSWERS - 180
Questions and Answers Already Graded A+ Premium Exam
Tested And Verified
Subject Area Applied Healthcare Statistics
Description This preassessment exam evaluates mastery of statistical methods applied to
healthcare data, including probability distributions, hypothesis testing, regression
analysis, and experimental design. It emphasizes interpretation of results and
application to clinical decision-making.
Expected Grade A+
Total Questions 180
Duration 3 hours
Learning Outcomes 1. Apply descriptive and inferential statistics to healthcare data
2. Interpret statistical results in clinical contexts
3. Design and analyze studies using appropriate statistical tests
4. Evaluate evidence-based practice using statistical reasoning
Accreditation This exam meets the rigor standards of top US R1 universities (e.g., Harvard,
Stanford) and accredited healthcare programs.
Page 1
,1. A healthcare researcher is analyzing the association between a new biomarker
(continuous) and disease progression (ordinal: mild, moderate, severe). Which
statistical method is most appropriate for assessing this association while adjusting
for confounding variables?
A. Pearson correlation coefficient
B. Spearman rank correlation
C. Ordinal logistic regression
D. Multinomial logistic regression
Answer: C. Ordinal logistic regression
Ordinal logistic regression (proportional odds model) is designed for an ordinal
outcome and can adjust for confounders. Spearman correlation only assesses bivariate
monotonic association without adjustment. Pearson requires continuous normal data.
Multinomial logistic regression treats outcome as nominal, ignoring ordering.
2. In a randomized controlled trial comparing two treatments, the primary outcome
is binary (success/failure). A logistic regression model yields an odds ratio of 1.5 for
treatment A vs B (p=0.03). Which interpretation is correct?
A. The relative risk of success is 1.5 for treatment A compared to B.
B. The odds of success are 1.5 times higher for treatment A than B, but this does not directly
equal relative risk.
C. Treatment A reduces the odds of failure by 50% relative to B.
D. The probability of success is 1.5 times greater for treatment A.
Answer: B. The odds of success are 1.5 times higher for treatment A than B, but
this does not directly equal relative risk.
Odds ratio approximates relative risk only when the outcome is rare. Here, without
incidence information, OR cannot be interpreted as risk ratio. Option B correctly
distinguishes odds from probability. Option D mistakes odds for probability. Option C
misstates direction (1.5 OR means higher odds of success, not lower failure).
Page 2
,3. A study reports a 95% confidence interval for the mean reduction in blood
pressure after a new drug: (-2.3, 5.1) mmHg. Which conclusion is most appropriate?
A. The drug significantly reduces blood pressure because the interval includes zero.
B. The drug does not have a statistically significant effect at =0.05 because the interval
includes zero.
C. The drug reduces blood pressure by an average of 2.3 to 5.1 mmHg with 95% confidence.
D. There is a 95% probability that the true mean reduction is between -2.3 and 5.1 mmHg.
Answer: B. The drug does not have a statistically significant effect at =0.05 because
the interval includes zero.
A confidence interval that includes zero indicates no statistically significant difference
at the corresponding level. Option A misinterprets inclusion of zero as significance.
Option C misstates the interval bounds. Option D is a common misinterpretation; the
correct interpretation is that 95% of such intervals contain the true mean.
4. In a survival analysis, the Kaplan-Meier estimator is used to compare survival
curves between two groups. The log-rank test yields a p-value of 0.04. Which
assumption is critical for the validity of this test?
A. The hazard ratio is constant over time.
B. Censoring is independent of the event time.
C. Survival times follow an exponential distribution.
D. The two groups have equal sample sizes.
Answer: B. Censoring is independent of the event time.
The log-rank test requires non-informative censoring (censoring independent of event
time). Option A is an assumption for Cox proportional hazards but not log-rank.
Option C is not required; KM is nonparametric. Option D is not necessary; unequal
sample sizes are allowed.
Page 3
, 5. A researcher plans a study to compare the means of three independent groups
using ANOVA. Preliminary data show that the variances are unequal (Levene's test
p=0.01) and the distributions are moderately skewed. Which approach is most
appropriate?
A. Proceed with standard one-way ANOVA because it is robust to violations.
B. Use Welch's ANOVA with a robust post-hoc test (e.g., Games-Howell).
C. Transform the data to achieve normality and equal variances, then use ANOVA.
D. Use a nonparametric Kruskal-Wallis test without further adjustment.
Answer: B. Use Welch's ANOVA with a robust post-hoc test (e.g., Games-Howell).
Welch's ANOVA does not assume equal variances and is more robust than standard
ANOVA when variances are unequal. Games-Howell post-hoc test also does not assume
equal variances. Transformations may not resolve both issues. Kruskal-Wallis is an
option but less powerful and does not directly compare means.
6. A diagnostic test has sensitivity of 95% and specificity of 90%. The prevalence of
the disease in the tested population is 2%. What is the positive predictive value
(PPV) of this test?
A. 0.162
B. 0.172
C. 0.190
D. 0.209
Answer: A. 0.162
PPV = (sensitivity × prevalence) / (sensitivity × prevalence +
(1-specificity)×(1-prevalence)). Plugging: (0.95×0.02)/(0.95×0.02 + 0.10×0.98) =
0.019/(0.019+0.098)=0.019/0.1170.1624. Thus 0.162.
Page 4
