Python Interview Questions for Data Analysts

Introduction

This comprehensive guide contains the most commonly asked Python interview questions in data analyst roles. The
questions cover essential topics including Pandas, NumPy, data manipulation, cleaning, and analysis—all critical skills for
data analyst interviews at major tech companies and organizations.

Interview Questions and Answers

1. What is Pandas, and why is it popular in data analysis?

Answer:
Pandas is an open-source Python library designed for data manipulation and analysis. It provides two primary data
structures:

Series: 1D labeled array (similar to a single column)

DataFrame: 2D labeled table (similar to a spreadsheet or SQL table)

Why it's popular:

Easy-to-use API for data cleaning and transformation

Handles missing data efficiently

Supports multiple data types (numeric, categorical, datetime)

Integrates well with other libraries (NumPy, Matplotlib, Scikit-learn)

Optimized performance for large datasets

Supports reading/writing various file formats (CSV, Excel, JSON, SQL databases)

Example:

import pandas as pd
df = pd.read_csv('[Link]')
print([Link]())

2. What is NumPy, and how is it different from Python lists?

Answer:
NumPy (Numerical Python) is a library for numerical computing that provides support for large, multi-dimensional arrays
and matrices, along with mathematical functions to operate on them efficiently.

Key Differences:

Feature Python Lists NumPy Arrays

Type Homogeneous or mixed Homogeneous only

Performance Slower Much faster (C implementation)

Memory Uses more memory Memory efficient

Operations Element-wise requires loops Vectorized operations

 Feature Python Lists NumPy Arrays

Ease of Use Simple syntax Requires learning NumPy functions

Mathematical Functions Limited Extensive (linear algebra, statistics)

Example:

import numpy as np

# Python list
python_list = [1, 2, 3, 4, 5]
result_list = [x * 2 for x in python_list] # Requires loop

# NumPy array
numpy_array = [Link]([1, 2, 3, 4, 5])
result_array = numpy_array * 2 # Vectorized operation (faster)

3. How do you read a CSV file into a Pandas DataFrame?

Answer:
You can read a CSV file using the pd.read_csv() function.

Syntax:

import pandas as pd

# Basic usage
df = pd.read_csv('[Link]')

# With additional parameters

df = pd.read_csv('[Link]',
sep=',', # Delimiter
header=0, # Row number for column names
dtype={'col1': int}, # Specify data types
nrows=100) # Read only first 100 rows

Common Parameters:

sep : Delimiter used in the file (default: ',')

header : Row number to use as column names (default: 0)

dtype : Dictionary specifying data types for columns

na_values : Additional strings to recognize as NaN

nrows : Number of rows to read

skiprows : Rows to skip

4. What is the difference between loc and iloc in Pandas?

Answer:
Both loc and iloc are used for indexing, but they work differently:

Feature loc iloc

Selection Type Label-based Integer position-based

 Feature loc iloc

Syntax loc[row_label, col_label] iloc[row_position, col_position]

Includes Both start and end are inclusive Start inclusive, end exclusive

Use Case When you know the label When you know the position

Examples:

import pandas as pd

df = [Link]({
'Name': ['Alice', 'Bob', 'Charlie'],
'Age': [25, 30, 35]
}, index=['A', 'B', 'C'])

# Using loc (label-based)

print([Link]['A', 'Name']) # Output: Alice
print([Link]['A':'B', 'Age']) # Both A and B included

# Using iloc (position-based)

print([Link][0, 0]) # Output: Alice
print([Link][0:2, 1]) # Rows 0-1 (2 excluded), column 1

5. How do you handle missing values in Pandas?

Answer:
There are several methods to handle missing values (NaN) in Pandas:

1. Remove missing values:

import pandas as pd

[Link]() # Remove rows with any NaN

[Link](subset=['col1']) # Remove rows with NaN in specific column
[Link](axis=1) # Remove columns with any NaN

2. Fill missing values:

[Link](0) # Fill with a specific value

[Link]([Link]()) # Fill with mean (numeric columns)
[Link](method='ffill') # Forward fill (use previous value)
[Link](method='bfill') # Backward fill (use next value)

3. Interpolate missing values:

[Link]() # Linear interpolation

[Link](method='bfill') # Backward interpolation

4. Check for missing values:

[Link]() # Returns boolean DataFrame

[Link]().sum() # Count missing values per column
6. Explain the concept of broadcasting in NumPy.
Answer:
Broadcasting is NumPy's mechanism for performing operations on arrays of different shapes. NumPy automatically
expands smaller arrays to match the shape of larger arrays without copying data, making operations efficient.

Broadcasting Rules:

1. If arrays have different numbers of dimensions, pad the smaller one with ones on the left

2. Arrays are compatible if dimensions are equal or one is 1

3. The array with dimension 1 is stretched to match the larger dimension

Examples:

import numpy as np

# Example 1: 1D array + scalar

arr = [Link]([1, 2, 3, 4])
result = arr + 10 # Output: [11, 12, 13, 14]

# Example 2: 2D + 1D broadcasting
matrix = [Link]([[1, 2, 3],
[4, 5, 6]]) # Shape: (2, 3)
vector = [Link]([1, 2, 3]) # Shape: (3,)
result = matrix + vector # Vector broadcasted to (2, 3)
# Output: [[2, 4, 6], [5, 7, 9]]

# Example 3: (2, 1) + (2, 3) broadcasting

col = [Link]([[10], [20]]) # Shape: (2, 1)
matrix = [Link]([[1, 2, 3],
[4, 5, 6]]) # Shape: (2, 3)
result = col + matrix # col broadcasted to (2, 3)

7. How do you filter rows in a DataFrame based on a condition?

Answer:
You can filter DataFrame rows using boolean indexing in Pandas.

Method 1: Single condition

import pandas as pd

df = [Link]({
'Name': ['Alice', 'Bob', 'Charlie'],
'Age': [25, 30, 35],
'Salary': [50000, 60000, 75000]
})

# Filter rows where Age > 25

filtered_df = df[df['Age'] > 25]

# Filter rows where Name == 'Bob'

filtered_df = df[df['Name'] == 'Bob']

Method 2: Multiple conditions (AND)

# Age > 25 AND Salary > 55000

filtered_df = df[(df['Age'] > 25) & (df['Salary'] > 55000)]
Method 3: Multiple conditions (OR)

# Age > 30 OR Salary > 70000

filtered_df = df[(df['Age'] > 30) | (df['Salary'] > 70000)]

Method 4: Using isin() for multiple values

# Filter where Name is either Alice or Charlie

filtered_df = df[df['Name'].isin(['Alice', 'Charlie'])]

Method 5: Using loc with conditions

filtered_df = [Link][df['Age'] > 25, ['Name', 'Age']]

8. How do you merge (join) two DataFrames in Pandas?

Answer:
Pandas provides merge() function to combine DataFrames based on a common column or index.

Syntax:

import pandas as pd

df1 = [Link]({
'ID': [1, 2, 3],
'Name': ['Alice', 'Bob', 'Charlie']
})

df2 = [Link]({
'ID': [1, 2, 3],
'Salary': [50000, 60000, 75000]
})

# Inner join (default)

merged_df = [Link](df1, df2, on='ID')

# Left join
merged_df = [Link](df1, df2, on='ID', how='left')

# Right join
merged_df = [Link](df1, df2, on='ID', how='right')

# Outer join
merged_df = [Link](df1, df2, on='ID', how='outer')

Join Types:

Join Type Description

inner Only matching rows from both DataFrames

left All rows from left DataFrame, matching rows from right

right All rows from right DataFrame, matching rows from left

outer All rows from both DataFrames

9. Explain the groupby() function in Pandas.
Answer:
groupby() is used to group rows by one or more columns and apply aggregate functions to each group.

Basic Syntax:

import pandas as pd

df = [Link]({
'Department': ['IT', 'IT', 'HR', 'HR', 'Sales'],
'Name': ['Alice', 'Bob', 'Charlie', 'Diana', 'Eve'],
'Salary': [50000, 60000, 45000, 55000, 40000]
})

# Group by Department and calculate mean salary

grouped = [Link]('Department')['Salary'].mean()
# Output:
# Department
# HR 50000.0
# IT 55000.0
# Sales 40000.0

# Multiple aggregations
grouped = [Link]('Department').agg({
'Salary': ['mean', 'sum', 'count'],
'Name': 'count'
})

# Multiple grouping columns

grouped = [Link](['Department', 'Name'])['Salary'].sum()

Common Aggregation Functions:

sum() : Sum of values

mean() : Average

median() : Median

min() : Minimum value

max() : Maximum value

count() : Count of values

std() : Standard deviation

var() : Variance

10. How do you handle duplicate rows in a DataFrame?

Answer:
Duplicate rows can be identified and removed using the duplicated() and drop_duplicates() functions.

Identifying duplicates:

import pandas as pd

df = [Link]({
'ID': [1, 2, 2, 3, 3, 3],
'Name': ['Alice', 'Bob', 'Bob', 'Charlie', 'Charlie', 'Charlie']
})
 # Check for duplicates
print([Link]()) # Returns boolean Series
print([Link](subset=['ID'])) # Check duplicates in specific column

Removing duplicates:

# Remove all duplicate rows

df_no_dups = df.drop_duplicates()

# Remove duplicates based on specific columns

df_no_dups = df.drop_duplicates(subset=['ID'], keep='first')

# Remove in-place
df.drop_duplicates(inplace=True)

Parameters:

subset : Column(s) to consider for identifying duplicates

keep : Which duplicates to keep ('first', 'last', or False for none)

inplace : Whether to modify the original DataFrame

11. What is the purpose of the apply() function in Pandas?

Answer:
apply() applies a function to each row, column, or element of a DataFrame or Series.

Using apply() on columns:

import pandas as pd

df = [Link]({
'A': [1, 2, 3],
'B': [4, 5, 6]
})

# Apply function to each column

result = [Link](sum) # Sums each column

# Apply custom function

result = [Link](lambda x: [Link]() - [Link]()) # Range for each column

Using apply() on rows:

# Apply function to each row (axis=1)

df['Total'] = [Link](lambda row: row['A'] + row['B'], axis=1)

# Create new column with conditional logic

df['Status'] = [Link](lambda row: 'High' if row['A'] > 2 else 'Low', axis=1)

Using apply() on Series:

# Apply to a Series
series = [Link]([1, 2, 3, 4, 5])
result = [Link](lambda x: x ** 2) # Square each value
12. How do you rename columns in a DataFrame?
Answer:
Columns can be renamed using the rename() method or by directly assigning to columns attribute.

Method 1: Using rename() with a dictionary

import pandas as pd

df = [Link]({
'col1': [1, 2, 3],
'col2': [4, 5, 6]
})

# Rename specific columns

df_renamed = [Link](columns={'col1': 'Column1', 'col2': 'Column2'})

# Rename in-place
[Link](columns={'col1': 'Column1'}, inplace=True)

Method 2: Assigning to columns attribute

[Link] = ['Column1', 'Column2']

Method 3: Using str methods

# Convert all column names to lowercase

[Link] = [Link]()

# Remove spaces from column names

[Link] = [Link](' ', '_')

13. What is the difference between Series and DataFrame in Pandas?

Answer:

Feature Series DataFrame

Dimensions 1D (single column) 2D (multiple rows and columns)

Structure Labeled array Table/spreadsheet-like structure

Axes Only index Index (rows) and columns

Creation [Link]() [Link]()

Use Case Single variable analysis Multi-variable analysis

Access series[index] df[column] or [Link][row]

Examples:

import pandas as pd

# Creating a Series
series = [Link]([10, 20, 30], index=['a', 'b', 'c'])
print(series['a']) # Output: 10
 # Creating a DataFrame
df = [Link]({
'A': [1, 2, 3],
'B': [4, 5, 6]
})
print(df['A']) # Returns a Series
print([Link][0]) # Returns a Series (row)

14. How do you calculate descriptive statistics for a DataFrame?

Answer:
Descriptive statistics summarize the central tendency, dispersion, and shape of a dataset.

Using describe():

import pandas as pd

df = [Link]({
'Age': [25, 30, 35, 40, 45],
'Salary': [50000, 60000, 75000, 80000, 95000]
})

# Get summary statistics

print([Link]())
# Output includes: count, mean, std, min, 25%, 50%, 75%, max

# For specific columns

print(df[['Age']].describe())

Individual statistics:

[Link]() # Mean
[Link]() # Median
[Link]() # Standard deviation
[Link]() # Variance
[Link]() # Minimum
[Link]() # Maximum
[Link]() # Sum
[Link]() # Count
[Link](0.25) # 25th percentile

15. How do you reshape a NumPy array?

Answer:
Reshaping changes the dimensions of an array without changing its data.

Using reshape():

import numpy as np

# 1D array to 2D
arr = [Link]([1, 2, 3, 4, 5, 6])
reshaped = [Link](2, 3)
# Output:
# [[1 2 3]
# [4 5 6]]

# 2D array to 3D
 arr_2d = [Link]([[1, 2, 3], [4, 5, 6]])
reshaped = arr_2d.reshape(2, 3, 1)

# Reshape with -1 (automatic dimension)

reshaped = [Link](2, -1) # NumPy calculates the -1 dimension

Using flatten() and ravel():

# Flatten to 1D
arr_2d = [Link]([[1, 2, 3], [4, 5, 6]])
flattened = arr_2d.flatten() # Returns copy: [1 2 3 4 5 6]
raveled = arr_2d.ravel() # Returns view (more efficient)

# Difference: flatten() returns a copy, ravel() returns a view

16. How do you create a pivot table in Pandas?

Answer:
A pivot table reorganizes data from rows to columns and applies aggregation functions.

Syntax:

import pandas as pd

df = [Link]({
'Date': ['2023-01-01', '2023-01-01', '2023-01-02', '2023-01-02'],
'Product': ['A', 'B', 'A', 'B'],
'Sales': [100, 200, 150, 250]
})

# Basic pivot table

pivot = pd.pivot_table(df,
values='Sales', # Values to aggregate
index='Date', # Row labels
columns='Product', # Column labels
aggfunc='sum') # Aggregation function

# Multiple aggregations
pivot = pd.pivot_table(df,
values='Sales',
index='Date',
columns='Product',
aggfunc=['sum', 'mean'])

Output example:

Product A B
Date
2023-01-01 100 200
2023-01-02 150 250
17. How do you calculate the correlation between variables in a DataFrame?
Answer:
Correlation measures the linear relationship between two variables, ranging from -1 to 1.

Calculating correlation:

import pandas as pd

df = [Link]({
'Age': [25, 30, 35, 40, 45],
'Salary': [50000, 60000, 75000, 80000, 95000]
})

# Correlation between two columns

corr = df['Age'].corr(df['Salary']) # Output: high positive correlation

# Correlation matrix (all numeric columns)

corr_matrix = [Link]()

# Correlation with a specific method

corr = [Link](method='pearson') # Pearson (default), spearman, kendall

Interpretation:

+1: Perfect positive correlation

0: No correlation

-1: Perfect negative correlation

18. How do you find the maximum and minimum values in a NumPy array?
Answer:
NumPy provides max() and min() functions for finding extreme values.

Finding max/min:

import numpy as np

arr = [Link]([3, 1, 4, 1, 5, 9, 2, 6])

# Maximum and minimum values

max_val = [Link]() # Output: 9
min_val = [Link]() # Output: 1

# 2D array operations
arr_2d = [Link]([[1, 2, 3], [4, 5, 6]])

# Max/min of entire array

max_val = arr_2d.max() # Output: 6

# Max/min along axis

max_per_col = arr_2d.max(axis=0) # Output: [4 5 6]
max_per_row = arr_2d.max(axis=1) # Output: [3 6]

# Finding index of max/min

max_index = [Link]() # Output: 5 (index where 9 is located)
min_index = [Link]() # Output: 1 (first index where 1 is located)
19. How do you create a new column in a DataFrame that is a transformation of existing columns?
Answer:
New columns can be created through various transformations of existing data.

Method 1: Direct calculation

import pandas as pd

df = [Link]({
'A': [1, 2, 3],
'B': [4, 5, 6]
})

# Create new column as sum of two columns

df['C'] = df['A'] + df['B']

# Create new column with scalar operation

df['D'] = df['A'] * 10

Method 2: Using apply() with lambda

# Create new column with conditional logic

df['Status'] = df['A'].apply(lambda x: 'High' if x > 2 else 'Low')

# Create new column using multiple columns

df['Ratio'] = [Link](lambda row: row['A'] / row['B'], axis=1)

Method 3: Using [Link]() for conditional assignment

import numpy as np

df['Category'] = [Link](df['A'] > 2, 'Above Average', 'Below Average')

Method 4: Using map() for categorical transformation

df['Grade'] = df['A'].map({1: 'F', 2: 'D', 3: 'C'})

20. How do you remove duplicate rows from a DataFrame?

Answer:
The drop_duplicates() method removes duplicate rows from a DataFrame.

Basic usage:

import pandas as pd

df = [Link]({
'Name': ['Alice', 'Bob', 'Alice', 'Charlie', 'Bob'],
'Age': [25, 30, 25, 35, 30]
})

# Remove all duplicates (based on all columns)

df_clean = df.drop_duplicates()

# Remove duplicates based on specific column(s)

df_clean = df.drop_duplicates(subset=['Name'])
 # Keep first occurrence (default)
df_clean = df.drop_duplicates(subset=['Name'], keep='first')

# Keep last occurrence

df_clean = df.drop_duplicates(subset=['Name'], keep='last')

# Remove all occurrences (keep=False)

df_clean = df.drop_duplicates(subset=['Name'], keep=False)

In-place removal:

df.drop_duplicates(inplace=True)

Quick Reference Table

Task Function Example

Read CSV pd.read_csv() df = pd.read_csv('[Link]')

Check shape .shape [Link] → (rows, columns)

First/last rows .head() , .tail() [Link](5)

Data types .dtypes [Link]

Missing values .isnull().sum() Counts NaN per column

Summary stats .describe() Mean, std, min, max, etc.

Sort data .sort_values() df.sort_values('Age')

Filter rows df[df['col'] > value] Boolean indexing

Group data .groupby() [Link]('col').sum()

Merge DataFrames [Link]() Inner/left/right/outer joins

Create column df['new'] = ... df['Total'] = df['A'] + df['B']

Remove duplicates .drop_duplicates() Removes duplicate rows

Replace values .replace() [Link](old_val, new_val)

Export to CSV .to_csv() df.to_csv('[Link]')

Tips for Interview Success

1. Practice with real datasets - Use Kaggle or real business data to practice
2. Know your libraries - Be comfortable with Pandas, NumPy, and basic Matplotlib

3. Understand SQL concepts - Many interview questions bridge SQL and Python
4. Write clean code - Use meaningful variable names and add comments
5. Explain your approach - Walk through your reasoning when solving problems
6. Test your code - Always verify output with small datasets first

7. Handle edge cases - Consider missing values, duplicates, and data type mismatches
8. Ask clarifying questions - Understand requirements before writing code
 9. Use vectorized operations - Avoid loops when possible for better performance

10. Review your logic - Double-check for off-by-one errors and incorrect assumptions