Exploring a Dataset

Learn a simple workflow for exploring any dataset in Pandas so you can understand its structure and spot potential issues before analysis.

Skill Level: Beginner
Prerequisites: Understanding DataFrames
Estimated Time: 25 minutes

Story Time

You now know how to load datasets into Pandas and how DataFrames are structured.

Before you start creating charts or building models, you need to understand what your dataset looks like. Professional data analysts begin by exploring the data: they check sample records, look at the size of the dataset, inspect data types, and review basic statistics.[web:4]

In this lesson, you will learn a simple workflow for exploring any dataset you load into Pandas, using the StudentsPerformance dataset as your example.[file:42]

What You’ll Learn

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

• Inspect the contents of a dataset.
• Understand the size and structure of a dataset.
• Identify data types.
• Generate summary statistics.
• Detect potential data quality issues before analysis.

Why This Topic Matters

It is risky to analyze a dataset that you do not understand.

Before running calculations or building visualizations, analysts typically ask:

• How many records are there?
• What variables are available?
• Which columns contain numbers?
• Are there missing values?
• Does the data appear reasonable?

This initial exploration helps you spot potential issues and guides the rest of your analysis.

A Typical Exploration Workflow

When working with a new dataset, it is helpful to follow a consistent exploratory pattern:

df.head()
df.info()
df.shape
df.describe()

These four commands provide a quick overview of the dataset:

• head() shows sample records.
• info() summarizes structure and types.
• shape shows the dimensions.
• describe() summarizes numerical columns.

Viewing Sample Records

Use head() to display the first few rows:

df.head()

By default, this shows the first five rows.

To show more rows:

df.head(10)

To view the last rows:

df.tail()

These previews help you confirm that the data loaded correctly and that the values look reasonable.

Understanding the Structure

Use info() to see a structural summary:

df.info()

This displays:

• number of rows
• number of columns
• column names
• data types
• information about missing values
• memory usage

It is often one of the first commands analysts run when they receive a new dataset.

Understanding the Size

Use shape to see the dimensions:

df.shape

Example output:

(1000, 8)

This means:

• 1000 rows (observations)
• 8 columns (variables)[file:42]

Summary Statistics

Use describe() to view summary statistics for numerical columns:

df.describe()

Example for a math score column:

Statistic	Math Score
Count	1000
Mean	66.09
Std	15.16
Min	0
Max	100

The output includes:

• count
• mean
• standard deviation
• minimum
• quartiles
• maximum

These statistics help you understand the typical values and spread of your numerical data.

Viewing Column Names

Use columns to list all variables:

df.columns

This returns the names of all columns and is useful when you start selecting or renaming variables.[file:42]

Counting Missing Values

Use isnull().sum() to count missing values per column:

df.isnull().sum()

For the StudentsPerformance dataset, each column should have zero missing values.[file:42]

Later in this guide, you will work with datasets that do contain missing values and learn how to handle them.

Worked Example

import pandas as pd

df = pd.read_csv("StudentsPerformance.csv")

df.head()
df.info()
df.describe()
df.shape

After running the code, answer:

• How many students are included?
• Which columns are numerical?
• Are there any missing values?
• What is the average math score?

These questions help you turn the exploratory output into useful understanding.[file:42]

Data Analysis Tip

Whenever you receive a new dataset, start with:

df.head()
df.info()
df.describe()
df.shape

Making this a habit will help you understand the structure of the data before performing any analysis.

Practice in the Notebook

Notebook

Open in Google Colab

Explore the StudentsPerformance dataset using Pandas.

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

In Google Colab:

1. Load the StudentsPerformance dataset into a DataFrame named df.

2. Run each of the following commands:

   df.head()
   df.tail()
   df.info()
   df.describe()
   df.shape
   df.columns
   df.isnull().sum()

3. Using the output, answer:

   • How many rows are in the dataset?
   • How many columns are present?
   • Which columns contain numerical values?
   • Which column has the highest average value?
   • Are there any missing values?

Record your answers before moving to the next lesson.

Independent Exercise

Load another dataset from the Dataset Library (or a different CSV file you have).

Without referring to your notes, investigate:

• number of rows
• number of columns
• column names
• numerical variables
• missing values

Summarize your observations in a few sentences.

Common Mistakes

Using head() Instead of describe()

Remember:

• head() displays sample records.
• describe() summarizes numerical data.

They serve different purposes.

Assuming All Columns Are Numerical

Datasets usually contain a mix of numerical and categorical variables. Always inspect data types (for example, using info()) before performing calculations.

Skipping Dataset Exploration

Jumping directly into analysis can lead to incorrect assumptions and misleading results. Spend a few minutes exploring every new dataset before writing analysis code.

Key Takeaways

In this lesson, you learned:

• How to inspect a dataset using head(), tail(), info(), describe(), shape, columns, and isnull().sum().
• How to understand the structure and size of a dataset.
• How to generate summary statistics for numerical columns.
• How to identify missing values and potential data quality issues.

Self Evaluation

Check Your Understanding

1. What does df.head() show?

The first few rows of the DataFrameSummary statistics for numerical columnsThe number of rows and columnsThe list of column names

2. Which command is commonly used to see summary statistics for numerical columns?

df.info()df.describe()df.head()df.isnull()

3. What does df.shape return?

The list of column namesThe number of missing valuesThe number of rows and columnsThe first and last rows

4. How can you check for missing values in each column?

df.head()df.info()df.isnull().sum()df.columns

5. Why should you explore a dataset before analysis?

To make the file smallerTo understand its structure and detect issues before running calculationsTo convert all values to textTo automatically build charts

Continue Your Journey

Next, you will learn how to select rows and columns, which is the first step in focusing on specific parts of your dataset during analysis.