Inspecting Data: head, tail, info, describe

Reviewed & published by Brayan K

Inspecting data means using Pandas' built-in summary methods — head, tail, info, and describe — to get a fast, reliable overview of a dataset's contents, structure, and statistics before you analyse it.

Part of the free Pandas course at LearnCodingFast — hands-on lessons with examples you run in your browser, plus practice exercises and a quick quiz.

Never trust data you haven't looked at. These methods are the first thing every analyst runs on a fresh dataset.

What You'll Learn in This Lesson

• Preview rows with head() and tail()
• Check structure and missing data with info()
• Summarise numbers with describe()
• Read size with shape, columns, dtypes
• Count distinct values with nunique()
• Tally categories with value_counts()

1 Previewing Rows: head() and tail()

Real datasets can have thousands of rows. df.head() shows the first 5 and df.tail() shows the last 5 — a quick peek without flooding your screen:

df.head()      # first 5 rows
df.head(3)     # first 3 rows
df.tail(2)     # last 2 rows

import pandas as pd

df = pd.DataFrame({
    "name": ["Ana", "Ben", "Cara", "Dan", "Eve", "Finn"],
    "score": [88, 72, 95, 60, 79, 84],
})

print(df.head(3))    # first 3 rows
print(df.tail(2))    # last 2 rows

# Output:
#   name  score
# 0  Ana     88
# 1  Ben     72
# 2 Cara     95
#   name  score
# 4  Eve     79
# 5 Finn     84

# ✅ Expected output:
#    name  score
# 0   Ana     88
# 1   Ben     72
# 2  Cara     95
#    name  score
# 4   Eve     79
# 5  Finn     84

2 Structure: info(), shape, columns, dtypes

df.info() prints a compact summary of the DataFrame's structure: how many rows, each column's name, its non-null count (great for spotting missing data), and its dtype.

• df.shape — (rows, columns)
• df.columns — the column names
• df.dtypes — the type of each column

import pandas as pd

df = pd.DataFrame({
    "name": ["Ana", "Ben", "Cara"],
    "score": [88, 72, 95],
    "passed": [True, True, True],
})

df.info()

# Output:
# <class 'pandas.core.frame.DataFrame'>
# RangeIndex: 3 entries, 0 to 2
# Data columns (total 3 columns):
#  #   Column  Non-Null Count  Dtype
# ---  ------  --------------  -----
#  0   name    3 non-null      object
#  1   score   3 non-null      int64
#  2   passed  3 non-null      bool
# dtypes: bool(1), int64(1), object(1)
# memory usage: 179.0+ bytes

# ✅ Expected output:
# <class 'pandas.core.frame.DataFrame'>
# RangeIndex: 3 entries, 0 to 2
# Data columns (total 3 columns):
#  #   Column  Non-Null Count  Dtype 
# ---  ------  --------------  ----- 
#  0   name    3 non-null      object
#  1   score   3 non-null      int64 
#  2   passed  3 non-null      bool  
# dtypes: bool(1), int64(1), object(1)
# memory usage: 183.0+ bytes

3 Statistics: describe(), nunique(), value_counts()

df.describe() computes summary statistics for every numeric column — count, mean, standard deviation, min, the three quartiles, and max:

df.describe()                 # numeric summary stats
df.nunique()                  # distinct values per column
df["city"].value_counts()     # tally of each category

nunique() tells you how many distinct values a column has, while value_counts() tallies how many times each value appears — ideal for category columns.

import pandas as pd

df = pd.DataFrame({
    "city": ["London", "Paris", "London", "Berlin", "Paris", "London"],
    "temp": [15, 18, 14, 12, 19, 16],
})

print(df["city"].value_counts())   # tally each city
print()
print("Distinct cities:", df["city"].nunique())
print("Mean temp:", df["temp"].mean())

# Output:
# city
# London    3
# Paris     2
# Berlin    1
# Name: count, dtype: int64
#
# Distinct cities: 3
# Mean temp: 15.666666666666666

# ✅ Expected output:
# city
# London    3
# Paris     2
# Berlin    1
# Name: count, dtype: int64
#
# Distinct cities: 3
# Mean temp: 15.666666666666666

import pandas as pd

df = pd.DataFrame({
    "name": ["Ana", "Ben", "Cara", "Dan"],
    "score": [88, 72, 95, 60],
})

# 🎯 YOUR TURN: print the numeric summary statistics
# Replace ___ with the right method name
print(df.___())

# Expected output (rows: count, mean, std, min, 25%, 50%, 75%, max):
#            score
# count   4.000000
# mean   78.750000
# std    15.305773
# min    60.000000
# 25%    69.000000
# 50%    80.000000
# 75%    89.750000
# max    95.000000

📋 Quick Reference

🎯 Mini Challenge: Quick Audit

Run a four-step audit on a small dataset:

• Print the shape
• Preview the first 2 rows with head(2)
• Print describe() for the numbers
• Tally the team column with value_counts()

import pandas as pd

df = pd.DataFrame({
    "player": ["Ana", "Ben", "Cara", "Dan"],
    "team": ["Red", "Blue", "Red", "Blue"],
    "points": [10, 7, 13, 9],
})

print("Shape:", df.shape)
print(df.head(2))
print(df["team"].value_counts())

# Output:
# Shape: (4, 3)
#   player  team  points
# 0    Ana   Red      10
# 1    Ben  Blue       7
# team
# Red     2
# Blue    2
# Name: count, dtype: int64

# ✅ Expected output:
# Shape: (4, 3)
#   player  team  points
# 0    Ana   Red      10
# 1    Ben  Blue       7
# team
# Red     2
# Blue    2
# Name: count, dtype: int64

❓ Frequently Asked Questions

Lesson complete — you can audit any dataset!

You can preview rows with head and tail, check structure and missing data with info, summarise numbers with describe, and tally categories with nunique and value_counts.

Practice quiz

Continue this course

• Previous: Reading & Writing Data (CSV, Excel, JSON)
• Next: Selecting Columns and Rows — Pick out the exact columns and rows you need from a DataFrame
• Quick reference: Pandas cheat sheet › Inspecting

Frequently asked questions

What does df.head() do in Pandas?

df.head() returns the first 5 rows of a DataFrame so you can preview its contents without printing everything. Pass a number to change how many, e.g. df.head(10). df.tail() does the same for the last rows.

What is the difference between info() and describe()?

df.info() summarises structure — column names, non-null counts, dtypes, and memory usage. df.describe() summarises the numbers — count, mean, std, min, quartiles, and max for each numeric column. One is about shape, the other about statistics.

How do I count unique values in a column?

Use df['col'].value_counts() to count how many times each distinct value appears (sorted high to low), or df.nunique() to get just the number of distinct values per column.

Which inspection method should I run first on a new dataset?

Start with df.shape and df.head() to see the size and a sample, then df.info() to check types and missing values, and df.describe() to understand the numeric ranges. Those four give you a fast, reliable overview.