Exploratory Data Analysis (EDA) template in Data Analysis Python - Step-by-Step Execution

Concept Flow - Exploratory Data Analysis (EDA) template

Load Data

↓

View Data Sample

↓

Check Data Info

↓

Summary Statistics

↓

Check Missing Values

↓

Visualize Data

↓

Draw Insights

This flow shows the main steps of EDA: loading data, viewing samples, checking info, summarizing stats, finding missing data, visualizing, and drawing insights.

Execution Sample

Data Analysis Python

import pandas as pd

df = pd.read_csv('data.csv')
print(df.head())
df.info()
print(df.describe())

This code loads a CSV file into a DataFrame, shows the first rows, data info, and summary statistics.

Execution Table

Step	Action	Code/Method	Output Example
1	Load Data	pd.read_csv('data.csv')	DataFrame with rows and columns loaded
2	View Data Sample	df.head()	First 5 rows of the DataFrame displayed
3	Check Data Info	df.info()	Shows column names, data types, non-null counts
4	Summary Statistics	df.describe()	Count, mean, std, min, max for numeric columns
5	Check Missing Values	df.isnull().sum()	Number of missing values per column
6	Visualize Data	df.hist() or sns.pairplot(df)	Histograms or pair plots to see distributions and relations
7	Draw Insights	Based on above outputs	Notes on data quality, trends, outliers
Exit	EDA Complete	All steps done	Ready for modeling or deeper analysis

💡 All EDA steps completed to understand data structure and quality

Variable Tracker

Variable	Start	After Load	After Head	After Info	After Describe	After Missing Check	After Visualization	Final
df	None	DataFrame loaded	First 5 rows shown	Info printed	Summary stats printed	Missing counts calculated	Plots generated	Insights noted

Key Moments - 3 Insights

Why do we look at df.head() before other steps?

df.head() shows a small sample of data so we can quickly see what columns and values look like before deeper checks (see execution_table step 2).

What does df.info() tell us that df.describe() does not?

df.info() shows data types and missing values count, while df.describe() only summarizes numeric columns (see execution_table steps 3 and 4).

Why check for missing values early in EDA?

Missing values affect analysis and modeling, so identifying them early (step 5) helps decide how to handle them.

Visual Quiz - 3 Questions

Look at the execution table, what output do we get from df.info() at step 3?

A Summary statistics like mean and std

B First 5 rows of data

C Data types and non-null counts per column

D Number of missing values per column

💡 Hint

Check the 'Output Example' column for step 3 in the execution_table

At which step do we calculate the number of missing values per column?

A Step 2

B Step 5

C Step 4

D Step 6

💡 Hint

Look for 'Check Missing Values' in the Action column of execution_table

If df.head() showed unexpected columns, which step would you repeat or check again?

A Step 1: Load Data

B Step 3: Check Data Info

C Step 6: Visualize Data

D Step 7: Draw Insights

💡 Hint

Unexpected columns usually mean data loaded incorrectly, so check step 1

Concept Snapshot

Exploratory Data Analysis (EDA) template:
1. Load data with pd.read_csv()
2. View sample rows with df.head()
3. Check data types and completeness with df.info()
4. Get summary stats with df.describe()
5. Find missing values with df.isnull().sum()
6. Visualize distributions and relations
7. Draw insights for next steps

Full Transcript

Exploratory Data Analysis (EDA) is a step-by-step process to understand data. First, we load the data into a DataFrame. Then, we look at a small sample using df.head() to see what the data looks like. Next, we check data types and missing values with df.info(). We get summary statistics for numeric columns using df.describe(). We count missing values with df.isnull().sum(). We visualize data distributions and relationships using plots. Finally, we draw insights about data quality and patterns to guide further analysis or modeling.

Practice

1. What is the main purpose of Exploratory Data Analysis (EDA) in data science?

easy

A. To build machine learning models directly

B. To clean data without checking it

C. To understand the data quickly and find patterns

D. To create final reports without analysis

Solution

1. Step 1: Understand EDA's role

   EDA is used to explore and understand data before any modeling or cleaning.

2. Step 2: Compare options

   Only To understand the data quickly and find patterns correctly states EDA's purpose as understanding data quickly and finding patterns.

3. Final Answer:

   To understand the data quickly and find patterns -> Option C

4. Quick Check:

   EDA = Understand data fast [OK]

Hint: EDA means exploring data first, not modeling [OK]

Common Mistakes:

• Confusing EDA with model building
• Thinking EDA skips data checking
• Assuming EDA creates final reports

2. Which Python command is commonly used to get a quick summary of a DataFrame's structure during EDA?

easy

A. df.describe()

B. df.info()

C. df.plot()

D. df.head()

Solution

1. Step 1: Recall DataFrame methods

   df.info() shows data types, non-null counts, and memory usage.

2. Step 2: Compare with other methods

   df.describe() gives stats, df.plot() makes graphs, df.head() shows rows. Only df.info() summarizes structure.

3. Final Answer:

   df.info() -> Option B

4. Quick Check:

   DataFrame structure = df.info() [OK]

Hint: Use df.info() for quick data structure summary [OK]

Common Mistakes:

• Using df.describe() for structure instead of stats
• Confusing df.head() with info summary
• Trying to plot before understanding data

3. Given the code:

import pandas as pd
df = pd.DataFrame({'A': [1, 2, 2, 3], 'B': [4, 5, 6, 7]})
print(df['A'].value_counts())

What is the output?

medium

A. 2 2
1 1
3 1
Name: A, dtype: int64

B. 1 2
2 1
3 1
Name: A, dtype: int64

C. 4 1
5 1
6 1
7 1
Name: B, dtype: int64

D. Error: value_counts() not found

Solution

1. Step 1: Understand value_counts()

   It counts how many times each unique value appears in the column.

2. Step 2: Count values in column 'A'

   Values are 1 (once), 2 (twice), 3 (once). value_counts() sorts by frequency descending, so: 2 2 1 1 3 1 Name: A, dtype: int64

3. Final Answer:

   2 2 1 1 3 1 Name: A, dtype: int64 -> Option A

4. Quick Check:

   Count unique values = 2 2 1 1 3 1 Name: A, dtype: int64 [OK]

Hint: value_counts() counts unique values in a column [OK]

Common Mistakes:

• Mixing counts of column B instead of A
• Misreading counts order
• Expecting error from value_counts()

4. You run this code:

import pandas as pd
df = pd.DataFrame({'X': [1, 2, None, 4], 'Y': [5, None, 7, 8]})
print(df.describe())

But the output misses some rows. What is the likely problem?

medium

A. describe() excludes rows with missing values by default

B. describe() ignores non-numeric columns

C. DataFrame has no numeric columns

D. Syntax error in code

Solution

1. Step 1: Understand describe() behavior

   describe() calculates stats on numeric columns but skips missing values in calculations.

2. Step 2: Check missing data effect

   Rows with None are ignored in count and stats, so output shows fewer rows counted.

3. Final Answer:

   describe() excludes rows with missing values by default -> Option A

4. Quick Check:

   Missing data skipped in describe() = describe() excludes rows with missing values by default [OK]

Hint: describe() skips missing values in stats by default [OK]

Common Mistakes:

• Thinking describe() drops entire rows
• Assuming syntax error causes missing rows
• Believing describe() works only on non-numeric data

5. You have a DataFrame df with columns age, income, and gender. You want to quickly check if income differs by gender. Which EDA step is best?

hard

A. Plot df['age'] histogram

B. Use df.describe() only

C. Drop missing values without checking

D. Use df.groupby('gender')['income'].mean() to compare averages

Solution

1. Step 1: Identify goal

   You want to compare income by gender groups.

2. Step 2: Choose correct EDA method

   Grouping by gender and calculating mean income shows differences clearly.

3. Final Answer:

   Use df.groupby('gender')['income'].mean() to compare averages -> Option D

4. Quick Check:

   Group and average to compare = Use df.groupby('gender')['income'].mean() to compare averages [OK]

Hint: Group by category and calculate mean to compare groups [OK]

Common Mistakes:

• Using describe() without grouping
• Plotting unrelated columns
• Dropping data before checking