WGU D600 - STATISTICAL DATA MINING TASK 2 | MULTIPLE
LOGISTIC REGRESSION ANALYSIS | COMPLETE
PERFORMANCE ASSESSMENT WITH PYTHON, CHURN
PREDICTION, CONFUSION MATRIX & ROC/AUC | 2026 UPDATED
147 Questions with Answers and Detailed Rationales
100 PERCENT GUARANTEED PASS
INSTANT DOWNLOAD ANSWERS INCLUDED
IMPORTANCE OF THIS DOCUMENT
This comprehensive examination preparation guide has been meticulously developed to help you succeed in the
WGU D600 - STATISTICAL DATA MINING TASK 2 | MULTIPLE LOGISTIC REGRESSION ANALYSIS |
COMPLETE PERFORMANCE ASSESSMENT WITH PYTHON, CHURN PREDICTION, CONFUSION MATRIX &
ROC/AUC | 2026 UPDATED. It contains 147 carefully selected questions that reflect the most current exam
content and testing strategies. Each question is accompanied by a correct answer and a detailed rationale that
explains the underlying pathophysiology, pharmacology, or clinical reasoning.
Self-Assessment – Test your knowledge and Exam Preparation – Familiarize yourself with the
identify areas requiring further question format and content
study areas
Concept Reinforcement – Deepen your Confidence Building – Develop test-taking
understanding through strategies and reduce
evidence-based exam anxiety
rationales
Time Management – Practice answering
questions under simulated
exam conditions
Review Summary 147 Questions
Foundations - Application - WGU D600 Statistical DATA Mining TASK 2 Multiple Logistic Regression
Analysis Complete Performance Assessment WITH Python Churn Prediction Confusion Matrix & Roc/auc
2026 Updated Statistical DATA Mining Multiple Logistic Regression Churn Prediction Model Evaluation
Confusion Matrix Roc/auc Undergraduate YEAR 3 / Upper-division WGU D600 Statistical DATA Mining
All answers with rationales
,Table of Contents
Content Area Questions Key Topics
DATA Preparation AND 1-25 Churn, Logistic, Model, Regression, Appropriate
Exploratory Analysis FOR
Churn Prediction
Simple AND Multiple Logistic 26-50 Churn, Model, Logistic, Regression, Multiple
Regression Model
Specification
Maximum Likelihood 51-75 Churn, Model, Logistic, Regression, Dataset
Estimation AND
Logit/log-odds Interpretation
ODDS Ratios Confidence 76-100 Churn, Model, Logistic Regression, Multiple Logistic, Coefficient
Intervals AND Statistical
Significance OF Predictors
Model FIT Assessment 101-125 Churn, Model, Logistic Regression, Positive, Class
Deviance Aic/bic AND
Pseudo R-squared Measures
Multicollinearity Interaction 126-147 Churn, Model, Logistic Regression, ODDS Ratio, Charges
Terms AND Variable
Selection Techniques
TOTAL 147 All questions include answers and detailed rationales
,Section A - DATA Preparation AND Exploratory Analysis
FOR Churn Prediction
Q1.
A logistic regression predicting customer churn yields a coefficient of 0.42 for
'contract_length_years.' Holding other predictors constant, what is the correct
interpretation of the odds ratio?
A. Each additional contract year multiplies B. Each additional contract year reduces the
the odds of churn by approximately 0.66. probability of churn by exactly 42
percentage points.
C. Each additional contract year increases D. The coefficient cannot be interpreted
the log-odds of churn by 0.42. without standardizing the predictor first.
Correct: A - Each additional contract year multiplies the odds of churn by approximately
0.66.
Rationale:The odds ratio is exp("0.42) "H 0.657, meaning each additional contract year
multiplies the odds of churn by ~0.66 (a ~34% reduction in odds). Option B confuses odds
with probability; C reverses the sign; D is false because logistic coefficients are interpretable
on the original scale.
Why the other answers are wrong:
B. Logistic coefficients affect odds (multiplicatively), not probability in percentage points.
C. The coefficient is negative, so it decreases-not increases-log-odds of churn.
D. Logistic regression coefficients are interpretable without standardization.
Reference: Hosmer, Lemeshow & Sturdivant (2024). Applied Logistic Regression, 4th Ed., Ch. 3.
Q2.
In a churn model, the confusion matrix shows TP=120, FP=80, FN=40, TN=760. Which
metric best reflects the model's ability to correctly identify actual churners?
A. Accuracy = 0.88 B. Precision = 0.60
C. Recall = 0.75 D. Specificity = 0.905
Correct: C - Recall = 0.75
Rationale:Recall (sensitivity) = TP/(TP+FN) = 120/160 = 0.75 and directly measures
detection of actual churners. Accuracy (0.88) is inflated by the majority class; precision (0.60)
measures correctness of positive predictions; specificity measures true negatives.
Why the other answers are wrong:
A. Accuracy is dominated by the majority non-churn class and does not isolate churn detection.
B. Precision = TP/(TP+FP) = 120/200 = 0.60, which measures positive predictive value, not
Page 3
, Section A - DATA Preparation AND Exploratory Analysis FOR Churn Prediction
detection of actual churners.
D. Specificity = TN/(TN+FP) = 760/840 0.905, measuring correct identification of non-churners.
Reference: James, Witten, Hastie & Tibshirani (2023). An Introduction to Statistical Learning, 2nd Ed.,
Ch. 4.
Q3.
You fit a logistic regression in statsmodels and observe a Variance Inflation Factor (VIF) of
12.4 for 'monthly_charge' and 9.8 for 'total_charges.' What is the most appropriate next
step?
A. Proceed-VIF values below 15 are always B. Drop the outcome variable and refit with a
acceptable in logistic models. linear regression.
C. Address multicollinearity by removing or D. Increase the sample size until VIF drops
combining one of the collinear predictors. below 5.
Correct: C - Address multicollinearity by removing or combining one of the collinear
predictors.
Rationale:VIF > 10 signals serious multicollinearity, which inflates standard errors and
destabilizes coefficient estimates. The standard remedy is to remove, combine, or apply
regularization (ridge) to collinear predictors. Increasing sample size does not directly reduce
VIF, which is a function of predictor correlations.
Why the other answers are wrong:
A. VIF > 10 indicates problematic multicollinearity; the 'below 15' claim is not a standard rule.
B. Changing the outcome type is unrelated to multicollinearity remediation.
D. VIF depends on predictor correlations, not sample size; adding data won't fix the structure.
Reference: Kutner, Nachtsheim & Neter (2024). Applied Linear Statistical Models, 6th Ed., Ch. 11.
Q4.
A churn classifier trained on a dataset with 5% churners achieves 95% accuracy by
predicting 'no churn' for everyone. Which evaluation approach would most clearly expose
this failure?
A. Report accuracy and overall error rate B. Report the ROC-AUC, which will be
only. approximately 0.5.
C. Report the mean squared error of D. Report the R-squared of the fitted model.
predictions.
Correct: B - Report the ROC-AUC, which will be approximately 0.5.
Rationale:A trivial classifier that always predicts the majority class produces an ROC curve
along the diagonal, yielding AUC 0.5, correctly signaling no discriminative power. Accuracy
(0.95) is misleading under class imbalance; MSE and R² are not standard classification
metrics.
Page 4
LOGISTIC REGRESSION ANALYSIS | COMPLETE
PERFORMANCE ASSESSMENT WITH PYTHON, CHURN
PREDICTION, CONFUSION MATRIX & ROC/AUC | 2026 UPDATED
147 Questions with Answers and Detailed Rationales
100 PERCENT GUARANTEED PASS
INSTANT DOWNLOAD ANSWERS INCLUDED
IMPORTANCE OF THIS DOCUMENT
This comprehensive examination preparation guide has been meticulously developed to help you succeed in the
WGU D600 - STATISTICAL DATA MINING TASK 2 | MULTIPLE LOGISTIC REGRESSION ANALYSIS |
COMPLETE PERFORMANCE ASSESSMENT WITH PYTHON, CHURN PREDICTION, CONFUSION MATRIX &
ROC/AUC | 2026 UPDATED. It contains 147 carefully selected questions that reflect the most current exam
content and testing strategies. Each question is accompanied by a correct answer and a detailed rationale that
explains the underlying pathophysiology, pharmacology, or clinical reasoning.
Self-Assessment – Test your knowledge and Exam Preparation – Familiarize yourself with the
identify areas requiring further question format and content
study areas
Concept Reinforcement – Deepen your Confidence Building – Develop test-taking
understanding through strategies and reduce
evidence-based exam anxiety
rationales
Time Management – Practice answering
questions under simulated
exam conditions
Review Summary 147 Questions
Foundations - Application - WGU D600 Statistical DATA Mining TASK 2 Multiple Logistic Regression
Analysis Complete Performance Assessment WITH Python Churn Prediction Confusion Matrix & Roc/auc
2026 Updated Statistical DATA Mining Multiple Logistic Regression Churn Prediction Model Evaluation
Confusion Matrix Roc/auc Undergraduate YEAR 3 / Upper-division WGU D600 Statistical DATA Mining
All answers with rationales
,Table of Contents
Content Area Questions Key Topics
DATA Preparation AND 1-25 Churn, Logistic, Model, Regression, Appropriate
Exploratory Analysis FOR
Churn Prediction
Simple AND Multiple Logistic 26-50 Churn, Model, Logistic, Regression, Multiple
Regression Model
Specification
Maximum Likelihood 51-75 Churn, Model, Logistic, Regression, Dataset
Estimation AND
Logit/log-odds Interpretation
ODDS Ratios Confidence 76-100 Churn, Model, Logistic Regression, Multiple Logistic, Coefficient
Intervals AND Statistical
Significance OF Predictors
Model FIT Assessment 101-125 Churn, Model, Logistic Regression, Positive, Class
Deviance Aic/bic AND
Pseudo R-squared Measures
Multicollinearity Interaction 126-147 Churn, Model, Logistic Regression, ODDS Ratio, Charges
Terms AND Variable
Selection Techniques
TOTAL 147 All questions include answers and detailed rationales
,Section A - DATA Preparation AND Exploratory Analysis
FOR Churn Prediction
Q1.
A logistic regression predicting customer churn yields a coefficient of 0.42 for
'contract_length_years.' Holding other predictors constant, what is the correct
interpretation of the odds ratio?
A. Each additional contract year multiplies B. Each additional contract year reduces the
the odds of churn by approximately 0.66. probability of churn by exactly 42
percentage points.
C. Each additional contract year increases D. The coefficient cannot be interpreted
the log-odds of churn by 0.42. without standardizing the predictor first.
Correct: A - Each additional contract year multiplies the odds of churn by approximately
0.66.
Rationale:The odds ratio is exp("0.42) "H 0.657, meaning each additional contract year
multiplies the odds of churn by ~0.66 (a ~34% reduction in odds). Option B confuses odds
with probability; C reverses the sign; D is false because logistic coefficients are interpretable
on the original scale.
Why the other answers are wrong:
B. Logistic coefficients affect odds (multiplicatively), not probability in percentage points.
C. The coefficient is negative, so it decreases-not increases-log-odds of churn.
D. Logistic regression coefficients are interpretable without standardization.
Reference: Hosmer, Lemeshow & Sturdivant (2024). Applied Logistic Regression, 4th Ed., Ch. 3.
Q2.
In a churn model, the confusion matrix shows TP=120, FP=80, FN=40, TN=760. Which
metric best reflects the model's ability to correctly identify actual churners?
A. Accuracy = 0.88 B. Precision = 0.60
C. Recall = 0.75 D. Specificity = 0.905
Correct: C - Recall = 0.75
Rationale:Recall (sensitivity) = TP/(TP+FN) = 120/160 = 0.75 and directly measures
detection of actual churners. Accuracy (0.88) is inflated by the majority class; precision (0.60)
measures correctness of positive predictions; specificity measures true negatives.
Why the other answers are wrong:
A. Accuracy is dominated by the majority non-churn class and does not isolate churn detection.
B. Precision = TP/(TP+FP) = 120/200 = 0.60, which measures positive predictive value, not
Page 3
, Section A - DATA Preparation AND Exploratory Analysis FOR Churn Prediction
detection of actual churners.
D. Specificity = TN/(TN+FP) = 760/840 0.905, measuring correct identification of non-churners.
Reference: James, Witten, Hastie & Tibshirani (2023). An Introduction to Statistical Learning, 2nd Ed.,
Ch. 4.
Q3.
You fit a logistic regression in statsmodels and observe a Variance Inflation Factor (VIF) of
12.4 for 'monthly_charge' and 9.8 for 'total_charges.' What is the most appropriate next
step?
A. Proceed-VIF values below 15 are always B. Drop the outcome variable and refit with a
acceptable in logistic models. linear regression.
C. Address multicollinearity by removing or D. Increase the sample size until VIF drops
combining one of the collinear predictors. below 5.
Correct: C - Address multicollinearity by removing or combining one of the collinear
predictors.
Rationale:VIF > 10 signals serious multicollinearity, which inflates standard errors and
destabilizes coefficient estimates. The standard remedy is to remove, combine, or apply
regularization (ridge) to collinear predictors. Increasing sample size does not directly reduce
VIF, which is a function of predictor correlations.
Why the other answers are wrong:
A. VIF > 10 indicates problematic multicollinearity; the 'below 15' claim is not a standard rule.
B. Changing the outcome type is unrelated to multicollinearity remediation.
D. VIF depends on predictor correlations, not sample size; adding data won't fix the structure.
Reference: Kutner, Nachtsheim & Neter (2024). Applied Linear Statistical Models, 6th Ed., Ch. 11.
Q4.
A churn classifier trained on a dataset with 5% churners achieves 95% accuracy by
predicting 'no churn' for everyone. Which evaluation approach would most clearly expose
this failure?
A. Report accuracy and overall error rate B. Report the ROC-AUC, which will be
only. approximately 0.5.
C. Report the mean squared error of D. Report the R-squared of the fitted model.
predictions.
Correct: B - Report the ROC-AUC, which will be approximately 0.5.
Rationale:A trivial classifier that always predicts the majority class produces an ROC curve
along the diagonal, yielding AUC 0.5, correctly signaling no discriminative power. Accuracy
(0.95) is misleading under class imbalance; MSE and R² are not standard classification
metrics.
Page 4
