Logistic Regression DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   How does logistic regression fundamentally differ from linear regression in terms of the outcome variable?

   • ALogistic regression uses a linear equation, while linear regression uses a nonlinear equation
   • BLogistic regression requires a normally distributed outcome, while linear regression does not
   • CLogistic regression predicts a categorical outcome, while linear regression predicts a continuous outcome
   • DLogistic regression does not use predictor variables, while linear regression does
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   What is the purpose of checking for multicollinearity in logistic regression?

   • ATo verify that the log-odds are linearly related to the predictors
   • BTo detect outliers in the outcome variable
   • CTo ensure the outcome variable is binary
   • DTo identify predictor variables that are highly correlated with each other
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A logistic regression model for employee attrition includes an extreme outlier: an employee with 50 years of tenure (data entry error). What is the most appropriate diagnostic step?

   • AKeep the outlier because logistic regression is robust to outliers
   • BRemove the outlier if it is a data entry error and document the decision
   • CIgnore the outlier because it only affects one data point
   • DTransform the outcome variable to reduce the outlier's impact
   Reveal Solution Discussion Select an answer first
5. 14application · easy

   A model's ROC curve is plotted. The curve is very close to the diagonal line. What does this indicate?

   • AThe model has poor discriminative ability, close to random guessing
   • BThe model has high precision but low recall
   • CThe model is overfitting the training data
   • DThe model has perfect discrimination
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A data scientist compares two logistic regression models for predicting patient readmission. Model A has AUC=0.85, Model B has AUC=0.78. Which conclusion is valid?

   • AModel A is better at ranking random positive cases higher than random negative cases
   • BModel B is overfitting because its AUC is lower
   • CModel A has higher accuracy than Model B
   • DModel A has a higher true positive rate at every threshold
   Reveal Solution Discussion Select an answer first