HarvardX: Data Science: Wrangling course Syllabus

Overview: This course provides a comprehensive introduction to data wrangling, a foundational skill in data science. You'll learn how to clean, transform, and combine real-world datasets into reliable, analysis-ready formats. The curriculum is structured into five core modules followed by a final project, requiring approximately 6–10 hours per week over 8–10 weeks. Through hands-on exercises, you’ll gain practical experience using widely adopted tools and techniques essential for data analysts and scientists.

Module 1: Introduction to Data Wrangling

Estimated time: 8 hours

• Understand the role of data wrangling in data science workflows
• Identify common data quality issues in real-world datasets
• Learn the principles of tidy data
• Explore structured data formats and their importance

Module 2: Data Cleaning and Transformation

Estimated time: 14 hours

• Handle missing data using appropriate strategies
• Detect and manage outliers in datasets
• Standardize inconsistent values and formats
• Apply transformations to improve data usability

Module 3: Data Manipulation and Reshaping

Estimated time: 14 hours

• Filter, sort, and summarize data efficiently
• Group data and compute aggregations
• Reshape datasets between wide and long formats
• Prepare data for visualization and analysis

Module 4: Working with Multiple Data Sources

Estimated time: 14 hours

• Combine datasets using joins and merges
• Understand relational data concepts
• Resolve key alignment and schema issues
• Prepare integrated datasets for downstream tasks

Module 5: Best Practices and Workflow Efficiency

Estimated time: 10 hours

• Document data cleaning steps reproducibly
• Use versioning and logging for transparency
• Optimize wrangling pipelines for performance

Module 6: Final Project

Estimated time: 20 hours

• Apply wrangling techniques to a messy real-world dataset
• Produce a clean, well-documented analysis-ready dataset
• Submit a short report explaining key cleaning and transformation decisions

Prerequisites

• Familiarity with basic programming concepts (e.g., variables, loops)
• Basic understanding of data tables and spreadsheet software
• Some prior exposure to data analysis or statistics is helpful

What You'll Be Able to Do After

• Identify and resolve common data quality issues
• Clean and standardize messy datasets systematically
• Transform and reshape data for analysis and visualization
• Combine multiple data sources using joins and merges
• Build reproducible, efficient data preparation workflows