MLS-C01 Question 115
Single answerYou are building a machine learning model to predict customer churn for an e-commerce platform. During exploratory data analysis, you calculate the correlation coefficients between your features and the target variable, 'Churn'. You find that 'Average Monthly Spend' has a correlation coefficient of -0.75 with 'Churn'. What can you infer from this correlation coefficient?
- A
There is a strong positive correlation between 'Average Monthly Spend' and 'Churn', meaning higher spenders are more likely to churn.
- B
There is a strong negative correlation between 'Average Monthly Spend' and 'Churn', meaning higher spenders are less likely to churn.
- C
There is no correlation between 'Average Monthly Spend' and 'Churn'.
- D
The p-value must be checked to confirm whether the correlation is statistically significant.
Show answer and explanation
Correct answer: B
Explanation
Correlation coefficients indicate the strength and direction of the linear relationship between two variables. A value of -0.75 shows a strong negative correlation, meaning that as 'Average Monthly Spend' increases, the target variable 'Churn' decreases. This insight is valuable for understanding the relationship between features and the target during exploratory data analysis.
- A. Incorrect.
Incorrect. The correlation coefficient of -0.75 indicates a strong negative, not positive, relationship between the variables.
- B. Correct.
Correct. A correlation coefficient of -0.75 indicates a strong negative relationship between 'Average Monthly Spend' and 'Churn'. This means that as 'Average Monthly Spend' increases, the likelihood of churn decreases.
- C. Incorrect.
Incorrect. A correlation coefficient of -0.75 clearly indicates a strong relationship, not no correlation.
- D. Incorrect.
Incorrect. While the p-value is useful to confirm statistical significance, the question is asking about the interpretation of the correlation coefficient itself, which shows a strong negative relationship.