IBM A1000-120 - Assessment: Data Science Foundations Intermediate Practice Exam: Medium Difficulty 2025

A data scientist is analyzing customer churn data and notices that 95% of customers stay with the company while only 5% churn. When building a classification model, the accuracy is 95% but the model fails to identify most churners. What is the primary issue and what metric would better evaluate this model?

A retail analyst is preparing sales data for visualization and notices that the 'revenue' column contains values ranging from $10 to $1,000,000 with a heavily right-skewed distribution. Which transformation would be most appropriate before creating visualizations to compare revenue across different product categories?

A data science team needs to determine if there is a statistically significant difference in conversion rates between three different website layouts (A: 12.5%, B: 14.2%, C: 13.8%) tested with equal sample sizes of 1000 visitors each. Which statistical test is most appropriate for this scenario?

During exploratory data analysis, a data scientist discovers that the 'age' feature has a correlation of 0.92 with a derived feature 'years_since_birth'. The dataset also includes 'income' and 'education_level' with a correlation of 0.78. What is the best approach to handle these relationships before model training?

A healthcare data scientist is building a model to predict patient readmission within 30 days. The dataset contains 50,000 records with 200 features. After initial training, the model achieves 98% accuracy on training data but only 75% on validation data. What combination of techniques would most effectively address this issue?

An e-commerce company wants to visualize the relationship between customer purchase frequency, average order value, and customer lifetime value across four different customer segments. Which visualization type would most effectively display these three continuous variables and four categories simultaneously?

A manufacturing company collects temperature sensor data every second, resulting in millions of records. The data science team needs to identify the central tendency for quality control purposes, but the data contains occasional extreme outliers due to sensor malfunctions. Which measure of central tendency should they use and why?

A data scientist is working with a dataset containing customer information where the 'income' column has 15% missing values. The data appears to be Missing Not At Random (MNAR) - higher income individuals are less likely to report income. What is the most appropriate strategy for handling these missing values?

A financial services company is comparing two predictive models for loan default. Model A has an AUC-ROC of 0.85 and predicts defaults with 80% precision and 70% recall. Model B has an AUC-ROC of 0.82 with 70% precision and 85% recall. The business priority is minimizing financial loss by catching as many potential defaults as possible. Which model should be selected and why?

A data analyst needs to present quarterly sales trends across 8 regions to executives who need to quickly identify underperforming regions and seasonal patterns. The data includes sales values, growth rates, and targets. Which combination of visualization techniques would be most effective?