Why Does Pandas Use Double Brackets: `df[[“area_m2”, “price_usd”]]`?

[Rajeev Bagra]

If you are learning Python Pandas for data analysis, you may have noticed something confusing:

Why do we sometimes write:

df[["area_m2", "price_usd"]]

instead of just:

df["area_m2", "price_usd"]

Why are there two sets of brackets?

In this post, we’ll explain this in a simple, beginner-friendly way.

---

✅ Understanding df[...] in Pandas

In Pandas, square brackets [] are used to select data from a DataFrame.

Think of it like this:

df[ ... ]

means:

“Get something from this DataFrame.”

It works like a dictionary:

my_dict["name"]

---

✅ Selecting a Single Column

When you want one column, you pass a string:

df["area_m2"]

What happens here?

• "area_m2" is a column name
• Pandas returns a Series (one column only)

Example:

type(df["area_m2"])
# pandas.core.series.Series

---

✅ Selecting Multiple Columns

When you want more than one column, you must use a list.

In Python, a list looks like this:

["area_m2", "price_usd"]

So in Pandas, you write:

df[["area_m2", "price_usd"]]

What does this mean?

• Inner [] → Python list of column names
• Outer [] → DataFrame selection

So it means:

“From df, select these columns.”

---

✅ Why Two Brackets Are Needed

Each bracket has a different role:

🔹 Outer Brackets → DataFrame Indexing

df[ ... ]

This tells Pandas:

“I want to access something from df.”

---

🔹 Inner Brackets → List of Columns

["area_m2", "price_usd"]

This tells Pandas:

“These are the columns I want.”

---

✅ Together:

df[["area_m2", "price_usd"]]

Means:

Select these columns from df.

---

✅ Equivalent Way (Easier to See)

You can also write:

columns = ["area_m2", "price_usd"]
df[columns]

This works the same way.

The double brackets are just a shortcut.

---

❌ Common Mistakes Beginners Make

❌ Mistake 1: Using Commas Without a List

df["area_m2", "price_usd"]   # ❌ Wrong

Why?

Python treats this as a tuple, not a list.
Pandas doesn’t accept this here.

---

❌ Mistake 2: Chaining Brackets

df["area_m2"]["price_usd"]   # ❌ Wrong

Why?

• First selects "area_m2"
• Then tries to find "price_usd" inside it
• It doesn’t exist

So it fails.

---

✅ Difference in Output: Series vs DataFrame

This is very important.

One Column → Series

df["area_m2"]

Returns:

pandas.Series

---

Multiple Columns → DataFrame

df[["area_m2", "price_usd"]]

Returns:

pandas.DataFrame

This matters because some methods work differently on Series and DataFrames.

---

✅ Practical Example

import pandas as pd

data = {
    "area_m2": [50, 70, 90],
    "price_usd": [100000, 150000, 200000],
    "city": ["Delhi", "Mumbai", "Pune"]
}

df = pd.DataFrame(data)

Select One Column

df["area_m2"]

Select Multiple Columns

df[["area_m2", "price_usd"]]

Get Summary Statistics

df[["area_m2", "price_usd"]].describe()

---

✅ Quick Rule to Remember

Task	Syntax
One column	df["col"]
Many columns	df[["col1", "col2"]]

👉 Rule:

Multiple columns = list = extra brackets

---

🧠 Why Pandas Uses This Design

Pandas follows standard Python logic:

• One item → string
• Many items → list

Instead of inventing new syntax, it uses normal Python data types.

This makes Pandas consistent and flexible.

---

🎯 Final Thoughts

The “double brackets” in Pandas are not special symbols.

They simply mean:

• Outer [] → Access DataFrame
• Inner [] → List of columns

Once you understand this, Pandas indexing becomes much easier.

---

[Author]

Posts