MLS-C01 Question 113
Select 2You are analyzing a dataset of customer transactions to determine which features are most predictive of the likelihood that a customer will churn. During your analysis, you compute the correlation coefficients between the numerical features and the target variable (churn) and observe the following results:
- Correlation between 'Total Spend' and 'Churn': -0.65
- Correlation between 'Customer Tenure' and 'Churn': 0.05
- Correlation between 'Number of Support Tickets' and 'Churn': 0.78
Based on these correlation values, which of the following statements are correct? (Select TWO answers.)
- A
A higher number of support tickets is strongly associated with an increased likelihood of churn.
- B
Total Spend is strongly negatively correlated with churn, meaning customers who spend more are less likely to churn.
- C
Customer Tenure has a strong positive correlation with churn.
- D
The correlation between Customer Tenure and Churn suggests no strong linear relationship.
- E
Correlation values alone are sufficient to conclude causation between features and churn.
Show answer and explanation
Correct answers: A, B
Explanation
Understanding correlation coefficients is crucial for interpreting relationships between features in a dataset. A high positive or negative correlation coefficient indicates a strong linear relationship, while low coefficients indicate weak or no linear relationships. However, correlation does not imply causation, and additional methods are required to determine causal relationships.
- A. Correct.
This is correct because the correlation value of 0.78 indicates a strong positive correlation. A higher number of support tickets is associated with an increased likelihood of churn.
- B. Correct.
This is correct because the correlation value of -0.65 indicates a strong negative correlation. Customers who spend more tend to have a lower likelihood of churning.
- C. Incorrect.
This is incorrect because the correlation value of 0.05 indicates a very weak or negligible positive correlation, not a strong one.
- D. Incorrect.
This is incorrect because while the correlation is weak, the statement does not directly answer the question about the relationship between features and churn.
- E. Incorrect.
This is incorrect because correlation does not imply causation. Other factors might influence the relationship, and further analysis is required to determine causation.