Selecting Rows and Columns

Learn how to select specific rows and columns in a Pandas DataFrame so you can focus on the parts of your dataset that matter for each analysis.

Skill Level: Beginner
Prerequisites: Exploring a Dataset
Estimated Time: 25 minutes

Story Time

The StudentsPerformance dataset contains many columns, but most analysis questions focus on just a few of them.[file:42]

For example, if you only want to study mathematics scores or compare all exam scores, you do not need every column in the DataFrame. Learning how to select specific rows and columns helps you work with just the information you need.

In this lesson, you will learn how to select columns and rows using Pandas.

What You’ll Learn

By the end of this lesson, you will be able to:

• Select one or more columns from a DataFrame.
• Select rows using labels and positions.
• Use .loc[] and .iloc[].
• Extract subsets of a dataset for analysis.

Why This Topic Matters

Most real analyses focus on subsets of the data.

You might want only exam scores, only students who took a test preparation course, or only records from a certain range of rows. If you can select specific rows and columns, you can:

• simplify your view of the data,
• reduce distraction from unrelated variables, and
• prepare focused datasets for each question.

Analysis Question

Suppose we want to answer:

What are the mathematics scores of all students?

To answer this question, we only need one column from the DataFrame.

Selecting a Single Column

Use square brackets with the column name:

df["math score"]

Example output:

0     72
1     69
2     90
...

Because only one column is selected, the result is a Series.

Selecting Multiple Columns

To compare all exam scores, select several columns at once:

df[["math score", "reading score", "writing score"]]

Example output:

Math	Reading	Writing
72	72	74
69	90	88
90	95	93

Notice the double square brackets: one set for the DataFrame selection, and one set for the list of column names.

Selecting Rows by Position with iloc

Use iloc[] when you want to select rows by their integer positions:

df.iloc

This returns the first row.

To select multiple rows:

df.iloc[0:5]

This returns rows from position 0 up to, but not including, position 5 (the first five rows).

Selecting Rows and Columns Together with iloc

You can select rows and columns at the same time:

df.iloc[0:5, 0:3]

This means:

• rows from position 0 to 4 (first five rows)
• columns from position 0 to 2 (first three columns)

Selecting Rows by Label with loc

Use loc[] when you want to select rows by their labels (index values):

df.loc

This selects the row whose index label is 0.

To select multiple rows:

df.loc[0:4]

With loc, both the start and end labels are included.

Selecting Specific Columns with loc

You can also select columns by name using loc:

df.loc[:, ["gender", "math score"]]

This means:

• : → all rows
• ["gender", "math score"] → only the gender and math score columns

Example output:

Gender	Math Score
Female	72
Female	69
Male	90

Difference Between loc and iloc

Function	Uses
loc[]	Labels (index, column names)
iloc[]	Integer positions

In practice:

• use loc[] when working with labels you recognize,
• use iloc[] when you want to slice by numeric positions.

Worked Example

Question:

Display only gender and mathematics score.

df[["gender", "math score"]]

Question:

Display the first ten students.

df.iloc[:10]

Question:

Display the first five students with only gender and reading score.

df.loc[:4, ["gender", "reading score"]]

These examples show how row and column selection can be combined.

Practice in the Notebook

Notebook

Open in Google Colab

Practice selecting rows and columns in the StudentsPerformance DataFrame.

After completing the activity in Google Colab, mark it as complete below.

Using the StudentsPerformance dataset in Colab, perform the following tasks:

1. Display only the mathematics score.
2. Display mathematics and reading scores.
3. Display the first ten students.
4. Display the last ten students.
5. Display only gender and writing score.

Try each selection in a separate code cell and observe the outputs.

Practice Exercise

Answer the following questions using Python:

1. Display only the parental level of education.
2. Display gender, lunch, and mathematics score.
3. Display the first fifteen students.
4. Display rows 50–60.
5. Display only reading and writing scores for the first twenty students.

Write the code for each task and run it to confirm your selections.

Independent Exercise

Suppose a university administrator wants to review only:

• gender
• test preparation course
• mathematics score

Write code to:

1. Display only these three columns.
2. Display the first twenty students with these columns.

This is a realistic example of preparing a focused view for a specific audience.

Common Mistakes

Using Single Brackets for Multiple Columns

Incorrect:

df["math score", "reading score"]

Correct:

df[["math score", "reading score"]]

Multiple column names must be provided as a list inside the brackets.

Confusing loc and iloc

Remember:

• loc[] uses labels (index values and column names).
• iloc[] uses integer positions.

Forgetting That Indexing Starts at Zero

df.iloc

returns the first row, not the second. Pandas uses zero‑based indexing.

Key Takeaways

In this lesson, you learned:

• How to select one or more columns from a DataFrame.
• How to select rows using loc[] and iloc[].
• How to combine row and column selections.
• How to extract subsets of a dataset for focused analysis.

Self Evaluation

Check Your Understanding

1. What does df["math score"] return?

A DataFrame with all columnsA Series containing only the math score columnA list of all column namesThe number of rows in the DataFrame

2. How do you select multiple columns: math, reading, and writing scores?

df["math score", "reading score", "writing score"]df[["math score", "reading score", "writing score"]]df.loc["math score", "reading score", "writing score"]df.iloc["math score", "reading score", "writing score"]

3. Which command selects the first five rows by position?

df.loc[0:4]df.iloc[0:5]df.head(5)df.loc[:5]

4. Which command uses labels to select all rows but only gender and math score?

df.iloc[:, ["gender", "math score"]]df.loc[:, ["gender", "math score"]]df["gender", "math score"]df.head()[["gender", "math score"]]

5. What is the main difference between loc[] and iloc[]?

loc[] uses labels; iloc[] uses integer positionsloc[] is only for columns; iloc[] is only for rowsloc[] works only with CSV files; iloc[] works only with Excel filesThere is no difference; they are identical

Continue Your Journey

Next, you will learn Filtering Data, where you’ll use conditions to select rows that meet specific criteria (such as students with scores above a certain threshold).