GroupBy Basics

The split-apply-combine pattern

GroupBy is one of pandas’ most powerful features. It follows a three-step pattern:

1. Split, divide the DataFrame into groups based on one or more columns
2. Apply, compute a function on each group independently
3. Combine, merge the results back into a single DataFrame

import pandas as pd

# titanic.csv ships with the course — load it from the browser file system.
df = pd.read_csv("titanic.csv")

Grouping by a single column

# Average survival rate by passenger class
print(df.groupby("Pclass")["Survived"].mean())

Output:

Pclass
1    0.629630
2    0.472826
3    0.242363
Name: Survived, dtype: float64

First-class passengers had a 63% survival rate, compared to 24% for third class. Groupby revealed a stark class difference in seconds.

What happens step by step:

# This is conceptually what groupby does:
for pclass, group_df in df.groupby("Pclass"):
    print(f"Class {pclass}: {group_df['Survived'].mean():.3f}")

Grouping by multiple columns

# Survival rate by class and sex
print(df.groupby(["Pclass", "Sex"])["Survived"].mean())

Output:

Pclass  Sex   
1       female    0.968085
        male      0.368852
2       female    0.921053
        male      0.157407
3       female    0.500000
        male      0.135447
Name: Survived, dtype: float64

Use unstack() to make this easier to read:

print(df.groupby(["Pclass", "Sex"])["Survived"].mean().unstack())

Aggregation methods

Groupby supports all standard aggregations:

# Mean fare by class
print(df.groupby("Pclass")["Fare"].mean())

# Total fare collected per class
print(df.groupby("Pclass")["Fare"].sum())

# Count of passengers per class
print(df.groupby("Pclass")["PassengerId"].count())

Multiple aggregations with agg()

The agg() method applies multiple functions at once:

print(df.groupby("Pclass")["Fare"].agg(["mean", "median", "min", "max", "count"]))

Output:

              mean  median     min       max  count
Pclass                                             
1        84.154687  60.287  0.0000  512.3292    216
2        20.662183  19.575  0.0000   73.5000    184
3        13.675550   8.050  0.0000   56.4958    491

Different aggregations per column:

print(df.groupby("Pclass").agg({
    "Survived": "mean",
    "Fare": ["mean", "max"],
    "Age": "median",
    "Name": "count"
}))

Aggregating all numeric columns

# Quick summary of all numeric columns per group
print(df.groupby("Pclass").mean(numeric_only=True))

GroupBy with filters

After grouping, you can filter entire groups:

# Keep only groups with more than 50 passengers
large_groups = df.groupby("Pclass").filter(lambda x: len(x) > 50)
print(large_groups["Pclass"].value_counts())

Try It

Using the Titanic dataset, calculate:

1. The average fare for each embarkation port
2. The survival rate for each combination of sex and embarkation port
3. The age statistics (mean, median, min, max) for each passenger class

import pandas as pd

# titanic.csv ships with the course — load it from the browser file system.
df = pd.read_csv("titanic.csv")

print("Average fare by port:")
print(df.groupby("Embarked")["Fare"].mean())

print("\nSurvival by sex and port:")
print(df.groupby(["Sex", "Embarked"])["Survived"].mean().unstack())

print("\nAge stats by class:")
print(df.groupby("Pclass")["Age"].agg(["mean", "median", "min", "max"]))

Key Takeaways

• GroupBy follows the split-apply-combine pattern: split data, apply a function, combine results
• Group by one column for simple summaries, multiple columns for deeper analysis
• agg() lets you compute multiple statistics at once, per column if needed
• Groupby reveals patterns that are invisible in the raw data

Practice Challenge

From the Titanic dataset, calculate the survival rate for each combination of Pclass, Sex, and whether the passenger was traveling alone (SibSp + Parch == 0). Which group had the highest survival rate? Which had the lowest?