Exercise

# Summary statistics for both classes

Consider the following `.groupby()`

code:

```
# Group by x and compute the standard deviation
df.groupby(['x']).std()
```

Here, a DataFrame `df`

is grouped by a column `'x'`

, and then the standard deviation is calculated across all columns of `df`

for each value of `'x'`

. The `.groupby()`

method is incredibly useful when you want to investigate specific columns of your dataset. Here, you're going to explore the `'Churn'`

column further to see if there are differences between churners and non-churners. A subset version of the `telco`

DataFrame, consisting of the columns `'Churn'`

, `'CustServ_Calls'`

, and `'Vmail_Message'`

is available in your workspace.

If you need a refresher on how `.groupby()`

works, please refer back to the pre-requisite Manipulating DataFrames with pandas course.

Instructions 1/3

**undefined XP**

Group `telco`

by `'Churn'`

and compute the mean.