Adding and removing columns in Data Analysis Python

Introduction

We add or remove columns to change the data we want to analyze. This helps us focus on important information or clean up the data.

You want to create a new column based on existing data, like calculating age from birth year.

You need to remove columns that are not useful or have too many missing values.

You want to add a column to label data, like marking sales as 'high' or 'low'.

You want to simplify the dataset by keeping only relevant columns for your analysis.

Syntax

Data Analysis Python

import pandas as pd

# Adding a column
df['new_column'] = values

# Removing a column
df = df.drop('column_name', axis=1)

Use axis=1 to specify you want to drop a column, not a row.

Adding a column can be done by assigning a list, a single value, or a calculation.

Examples

This adds a new column 'C' which is the sum of columns 'A' and 'B'.

Data Analysis Python

import pandas as pd

df = pd.DataFrame({'A': [1, 2], 'B': [3, 4]})
# Add column C as sum of A and B
df['C'] = df['A'] + df['B']

This removes the column 'B' from the DataFrame.

Data Analysis Python

import pandas as pd

df = pd.DataFrame({'A': [1, 2], 'B': [3, 4]})
# Remove column B
df = df.drop('B', axis=1)

This adds a new column 'constant' with value 10 for every row.

Data Analysis Python

import pandas as pd

df = pd.DataFrame({'A': [1, 2]})
# Add a column with the same value for all rows
df['constant'] = 10

Sample Program

This program creates a table with names, ages, and salaries. It adds a new column showing age after 5 years. Then it removes the salary column. Finally, it prints the updated table.

Data Analysis Python

import pandas as pd

# Create a simple DataFrame
df = pd.DataFrame({
    'Name': ['Alice', 'Bob', 'Charlie'],
    'Age': [25, 30, 35],
    'Salary': [50000, 60000, 70000]
})

# Add a new column 'Age in 5 years'
df['Age in 5 years'] = df['Age'] + 5

# Remove the 'Salary' column
df = df.drop('Salary', axis=1)

print(df)

Important Notes

Dropping columns does not change the original DataFrame unless you assign it back or use inplace=True.

Adding columns with calculations helps create new insights from existing data.

Summary

You add columns to include new information or calculations.

You remove columns to clean or simplify your data.

Use df['new_col'] = ... to add and df.drop(..., axis=1) to remove columns.

Practice

1. What is the correct way to add a new column named 'age' with all values set to 30 in a pandas DataFrame df?

easy

A. df.append('age', 30)

B. df.add_column('age', 30)

C. df['age'] = 30

D. df.insert_column('age', 30)

Solution

1. Step 1: Understand how to add columns in pandas

   In pandas, you add a new column by assigning a value or list to df['column_name'].

2. Step 2: Apply the correct syntax to add a column named 'age'

   Using df['age'] = 30 adds a new column 'age' with all values set to 30.

3. Final Answer:

   df['age'] = 30 -> Option C

4. Quick Check:

   Adding column = df['new_col'] = value [OK]

Hint: Use df['col_name'] = value to add columns fast [OK]

Common Mistakes:

• Using non-existent methods like add_column or insert_column
• Trying to use append which is for rows, not columns
• Forgetting to use square brackets for column name

2. Which of the following is the correct syntax to remove a column named 'salary' from a pandas DataFrame df?

easy

A. df.drop('salary', axis=1)

B. df.remove('salary')

C. df.delete_column('salary')

D. df.drop_column('salary')

Solution

1. Step 1: Recall pandas method to remove columns

   The pandas method to remove columns is drop() with axis=1 to specify columns.

2. Step 2: Use correct syntax to drop 'salary' column

   Using df.drop('salary', axis=1) removes the 'salary' column from the DataFrame.

3. Final Answer:

   df.drop('salary', axis=1) -> Option A

4. Quick Check:

   Remove column = df.drop('col', axis=1) [OK]

Hint: Use df.drop('col', axis=1) to remove columns [OK]

Common Mistakes:

• Using methods like remove or delete_column which don't exist
• Forgetting axis=1 to specify column removal
• Using drop without parentheses or wrong parameters

3. Given the DataFrame df below:

   name  score
0  Amy     90
1  Bob     80
2  Cara    85

What will be the output of this code?

df['passed'] = df['score'] >= 85
print(df)

medium

A. name score passed
0 Amy 90 False
1 Bob 80 True
2 Cara 85 False

B. name score passed
0 Amy 90 True
1 Bob 80 False
2 Cara 85 True

C. name score
0 Amy 90
1 Bob 80
2 Cara 85

D. name score passed
0 Amy 90 90
1 Bob 80 80
2 Cara 85 85

Solution

1. Step 1: Understand the condition applied to 'score'

   The expression df['score'] >= 85 creates a boolean series: True where score is 85 or more, False otherwise.

2. Step 2: Assign boolean series to new column 'passed'

   Adding df['passed'] = ... adds this boolean series as a new column.

3. Final Answer:

   DataFrame with 'passed' column showing True for scores >= 85 -> Option B

4. Quick Check:

   Boolean condition adds True/False column [OK]

Hint: Compare column with condition to add True/False column [OK]

Common Mistakes:

• Expecting numeric values instead of booleans in new column
• Not using >= operator correctly
• Printing df before adding the new column

4. What is wrong with this code snippet to remove the column 'height' from DataFrame df?

df.drop('height')

medium

A. Missing axis=1 to specify column removal

B. The column name should be in a list like ['height']

C. The method drop does not exist in pandas

D. You must assign the result back to df like df = df.drop('height')

Solution

1. Step 1: Check the drop method parameters

   By default, drop() removes rows. To remove columns, you must specify axis=1.

2. Step 2: Identify the missing axis parameter

   The code misses axis=1, so it tries to remove a row named 'height', causing an error if no such row exists.

3. Final Answer:

   Missing axis=1 to specify column removal -> Option A

4. Quick Check:

   drop columns = drop('col', axis=1) [OK]

Hint: Always add axis=1 to drop columns, not rows [OK]

Common Mistakes:

• Forgetting axis=1 causes wrong row removal or error
• Thinking drop modifies df inplace without assignment
• Using wrong data type for column name

5. You have a DataFrame df with columns ['A', 'B', 'C']. You want to create a new column 'D' which is the sum of 'A' and 'B', then remove column 'C'. Which code correctly does this in one step?

hard

A. df['D'] = df['A'] + df['B']
df.remove('C')

B. df['D'] = df['A'] + df['B']
df.drop('C', axis=1, inplace=True)

C. df['D'] = df['A'] + df['B']
df = df.drop('C', axis=1)

D. df = df.assign(D=df['A'] + df['B']).drop('C', axis=1)

Solution

1. Step 1: Create new column 'D' as sum of 'A' and 'B'

   Using assign() allows adding a new column in a chainable way without modifying df inplace.

2. Step 2: Remove column 'C' using drop() chained after assign()

   Chaining drop('C', axis=1) removes column 'C' and returns the modified DataFrame.

3. Step 3: Assign the result back to df

   Assigning the chained operations back to df updates the DataFrame correctly in one step.

4. Final Answer:

   df = df.assign(D=df['A'] + df['B']).drop('C', axis=1) -> Option D

5. Quick Check:

   Use assign + drop chained and assign back [OK]

Hint: Chain assign and drop, then assign back to df [OK]

Common Mistakes:

• Using drop without axis=1 or inplace causing no change
• Using non-existent remove method
• Not assigning chained result back to df