,WGU C797 Data Science & Analytics Advanced Prep:
Master Statistical Inference & Data Modeling Practice
Questions & Detailed Explanations
Subject: C797 Data Science and Analytics - Statistical Inference, Modeling, and
Healthcare Informatics
Question 1: A healthcare researcher is conducting a study to determine if a new telehealth
intervention significantly reduces patient readmission rates. The researcher uses a randomized
control trial design. What is the primary purpose of the randomization process in this study?
A) To ensure that the sample size is statistically sufficient to meet power requirements.
B) To eliminate selection bias by ensuring that both known and unknown confounding variables
are distributed equally between the intervention and control groups.
C) To allow the researcher to manipulate the independent variable more effectively.
D) To guarantee that the results are generalizable to the entire global population.
Correct Answer: B) To eliminate selection bias by ensuring that both known and unknown
confounding variables are distributed equally between the intervention and control groups.
Explanation: Randomization is the hallmark of a true experiment because it minimizes selection
bias. By randomly assigning participants, researchers balance baseline characteristics
(confounders) that could otherwise skew results. Option A refers to power analysis, not
randomization; C refers to the nature of experimental design; and D is limited by the sampling
frame, not randomization.
Question 2: In a cross-sectional study analyzing the relationship between EHR documentation
speed and physician burnout, the researcher identifies a correlation coefficient ($r$) of -0.75.
Which of the following is the most appropriate interpretation of this result?
A) There is a weak positive relationship between documentation speed and burnout.
B) Increased documentation speed causes a decrease in physician burnout.
C) There is a strong negative association where faster documentation speed is associated with
lower levels of burnout.
D) The data suggests that 75% of the variation in burnout is explained by documentation speed.
Correct Answer: C) There is a strong negative association where faster documentation speed
is associated with lower levels of burnout.
,Explanation: A correlation coefficient of -0.75 indicates a strong inverse relationship. Because
cross-sectional studies are observational, they cannot definitively prove causation (eliminating
B). Option D is incorrect because the coefficient of determination ($R^2$) would be $(-0.75)^2 =
0.5625$, not 0.75.
Question 3: A data analyst is performing a chi-square test to determine if there is a significant
association between patient ethnicity and the utilization of a specific preventative health service.
The calculated p-value is 0.035. Using a standard alpha level of 0.05, what is the correct
conclusion?
A) Fail to reject the null hypothesis; there is insufficient evidence of an association.
B) Reject the null hypothesis; there is statistically significant evidence of an association.
C) The null hypothesis is proven true.
D) The test is invalid because chi-square cannot be used for categorical ethnicity data.
Correct Answer: B) Reject the null hypothesis; there is statistically significant evidence of an
association.
Explanation: When the p-value (0.035) is less than the significance level (alpha = 0.05), the null
hypothesis—which posits no association—is rejected in favor of the alternative hypothesis. Chi-
square is precisely the correct test for comparing categorical variables like ethnicity and service
usage.
Question 4: Which of the following best describes the Central Limit Theorem (CLT) in the
context of data analytics?
A) As the sample size increases, the sample mean will eventually equal the exact population
mean.
B) The distribution of sample means will approximate a normal distribution, regardless of the
population distribution, as the sample size becomes sufficiently large.
C) The standard deviation of the sample will always equal the standard deviation of the
population.
D) Data must be normally distributed in the population for inferential statistics to be applicable.
Correct Answer: B) The distribution of sample means will approximate a normal
distribution, regardless of the population distribution, as the sample size becomes
sufficiently large.
, Explanation: The CLT is foundational for inferential statistics. It allows researchers to use
normal distribution-based tests (like t-tests) even when the underlying data is not normal,
provided the sample size is large enough (typically $n \ge 30$).
Question 5: A clinical system measures the diagnostic sensitivity of a new blood test for sepsis.
If the sensitivity is 0.90, what does this value signify?
A) 90% of the patients who do not have sepsis will test negative.
B) 90% of the patients who have sepsis will correctly test positive.
C) 90% of all positive tests are true positives.
D) 90% of the population tested does not have the disease.
Correct Answer: B) 90% of the patients who have sepsis will correctly test positive.
Explanation: Sensitivity is defined as the probability that a test result will be positive among
those who actually have the disease (True Positives / [True Positives + False Negatives]).
Option A describes specificity.
Question 6: In a logistic regression model predicting hospital readmission, the odds ratio (OR)
for a specific comorbid condition is 2.5 (95% CI: 1.8 – 3.2). How should this be interpreted?
A) Patients with the condition have 2.5 times the odds of being readmitted compared to those
without the condition.
B) The condition increases the risk of readmission by 25%.
C) The condition reduces the odds of readmission by 2.5 times.
D) Because the confidence interval does not include 0, the result is non-significant.
Correct Answer: A) Patients with the condition have 2.5 times the odds of being readmitted
compared to those without the condition.
Explanation: An OR of 2.5 indicates that the event (readmission) is 2.5 times more likely to occur
in the presence of the exposure (comorbidity). The 95% CI (1.8–3.2) does not include 1.0, which
indicates the result is statistically significant.
Question 7: A nursing informaticist is evaluating the "Efficiency" of a diagnostic tool. In clinical
informatics, what does test efficiency represent?
A) The probability that the test result (positive or negative) and the clinical diagnosis agree.
B) The time it takes for the laboratory to process the test.
Master Statistical Inference & Data Modeling Practice
Questions & Detailed Explanations
Subject: C797 Data Science and Analytics - Statistical Inference, Modeling, and
Healthcare Informatics
Question 1: A healthcare researcher is conducting a study to determine if a new telehealth
intervention significantly reduces patient readmission rates. The researcher uses a randomized
control trial design. What is the primary purpose of the randomization process in this study?
A) To ensure that the sample size is statistically sufficient to meet power requirements.
B) To eliminate selection bias by ensuring that both known and unknown confounding variables
are distributed equally between the intervention and control groups.
C) To allow the researcher to manipulate the independent variable more effectively.
D) To guarantee that the results are generalizable to the entire global population.
Correct Answer: B) To eliminate selection bias by ensuring that both known and unknown
confounding variables are distributed equally between the intervention and control groups.
Explanation: Randomization is the hallmark of a true experiment because it minimizes selection
bias. By randomly assigning participants, researchers balance baseline characteristics
(confounders) that could otherwise skew results. Option A refers to power analysis, not
randomization; C refers to the nature of experimental design; and D is limited by the sampling
frame, not randomization.
Question 2: In a cross-sectional study analyzing the relationship between EHR documentation
speed and physician burnout, the researcher identifies a correlation coefficient ($r$) of -0.75.
Which of the following is the most appropriate interpretation of this result?
A) There is a weak positive relationship between documentation speed and burnout.
B) Increased documentation speed causes a decrease in physician burnout.
C) There is a strong negative association where faster documentation speed is associated with
lower levels of burnout.
D) The data suggests that 75% of the variation in burnout is explained by documentation speed.
Correct Answer: C) There is a strong negative association where faster documentation speed
is associated with lower levels of burnout.
,Explanation: A correlation coefficient of -0.75 indicates a strong inverse relationship. Because
cross-sectional studies are observational, they cannot definitively prove causation (eliminating
B). Option D is incorrect because the coefficient of determination ($R^2$) would be $(-0.75)^2 =
0.5625$, not 0.75.
Question 3: A data analyst is performing a chi-square test to determine if there is a significant
association between patient ethnicity and the utilization of a specific preventative health service.
The calculated p-value is 0.035. Using a standard alpha level of 0.05, what is the correct
conclusion?
A) Fail to reject the null hypothesis; there is insufficient evidence of an association.
B) Reject the null hypothesis; there is statistically significant evidence of an association.
C) The null hypothesis is proven true.
D) The test is invalid because chi-square cannot be used for categorical ethnicity data.
Correct Answer: B) Reject the null hypothesis; there is statistically significant evidence of an
association.
Explanation: When the p-value (0.035) is less than the significance level (alpha = 0.05), the null
hypothesis—which posits no association—is rejected in favor of the alternative hypothesis. Chi-
square is precisely the correct test for comparing categorical variables like ethnicity and service
usage.
Question 4: Which of the following best describes the Central Limit Theorem (CLT) in the
context of data analytics?
A) As the sample size increases, the sample mean will eventually equal the exact population
mean.
B) The distribution of sample means will approximate a normal distribution, regardless of the
population distribution, as the sample size becomes sufficiently large.
C) The standard deviation of the sample will always equal the standard deviation of the
population.
D) Data must be normally distributed in the population for inferential statistics to be applicable.
Correct Answer: B) The distribution of sample means will approximate a normal
distribution, regardless of the population distribution, as the sample size becomes
sufficiently large.
, Explanation: The CLT is foundational for inferential statistics. It allows researchers to use
normal distribution-based tests (like t-tests) even when the underlying data is not normal,
provided the sample size is large enough (typically $n \ge 30$).
Question 5: A clinical system measures the diagnostic sensitivity of a new blood test for sepsis.
If the sensitivity is 0.90, what does this value signify?
A) 90% of the patients who do not have sepsis will test negative.
B) 90% of the patients who have sepsis will correctly test positive.
C) 90% of all positive tests are true positives.
D) 90% of the population tested does not have the disease.
Correct Answer: B) 90% of the patients who have sepsis will correctly test positive.
Explanation: Sensitivity is defined as the probability that a test result will be positive among
those who actually have the disease (True Positives / [True Positives + False Negatives]).
Option A describes specificity.
Question 6: In a logistic regression model predicting hospital readmission, the odds ratio (OR)
for a specific comorbid condition is 2.5 (95% CI: 1.8 – 3.2). How should this be interpreted?
A) Patients with the condition have 2.5 times the odds of being readmitted compared to those
without the condition.
B) The condition increases the risk of readmission by 25%.
C) The condition reduces the odds of readmission by 2.5 times.
D) Because the confidence interval does not include 0, the result is non-significant.
Correct Answer: A) Patients with the condition have 2.5 times the odds of being readmitted
compared to those without the condition.
Explanation: An OR of 2.5 indicates that the event (readmission) is 2.5 times more likely to occur
in the presence of the exposure (comorbidity). The 95% CI (1.8–3.2) does not include 1.0, which
indicates the result is statistically significant.
Question 7: A nursing informaticist is evaluating the "Efficiency" of a diagnostic tool. In clinical
informatics, what does test efficiency represent?
A) The probability that the test result (positive or negative) and the clinical diagnosis agree.
B) The time it takes for the laboratory to process the test.
