Complete eight data science projects that lock in important real-world skills—along with a practical process you can use to learn any new technique quickly and efficiently.

Data analysts need to be problem solvers—and The Well-Grounded Data Analyst will teach you how to solve the most common problems you'll face in industry. You'll explore eight scenarios that your class or bootcamp won’t have covered, so you can accomplish what your boss is asking for.

In The Well-Grounded Data Analyst you'll learn:

• High-value skills to tackle specific analytical problems

• Deconstructing problems for faster, practical solutions

• Data modeling, PDF data extraction, and categorical data manipulation

• Handling vague metrics, deciphering inherited projects, and defining customer records

The Well-Grounded Data Analyst is for junior and early-career data analysts looking to supplement their foundational data skills with real-world problem solving. As you explore each project, you'll also master a proven process for quickly learning new skills developed by author and Half Stack Data Science podcast host David Asboth. You'll learn how to determine a minimum viable answer for your stakeholders, identify and obtain the data you need to deliver, and reliably present and iterate on your findings. The book can be read cover-to-cover or opened to the chapter most relevant to your current challenges.

Foreword by Reuven M. Lerner.

About the technology

Real world data analysis is messy. Success requires tackling challenges like unreliable data sources, ambiguous requests, and incompatible formats—often with limited guidance. This book goes beyond the clean, structured examples you see in classrooms and bootcamps, offering a step-by-step framework you can use to confidently solve any data analysis problem like a pro.

About the book

The Well-Grounded Data Analyst introduces you to eight scenarios that every data analyst is bound to face. You’ll practice author David Asboth’s results-oriented approach as you model data by identifying customer records, navigate poorly-defined metrics, extract data from PDFs, and much more! It also teaches you how to take over incomplete projects and create rapid prototypes with real data. Along the way, you’ll build an impressive portfolio of projects you can showcase at your next interview.

What's inside

• Deconstructing problems

• Handling vague metrics

• Data modeling

• Categorical data manipulation

About the reader

For early-career data scientists.

About the author

David Asboth is a data generalist educator, and software architect. He co-hosts the Half Stack Data Science podcast.

Table of Contents

1 Bridging the gap between data science training and the real world

2 Encoding geographies

3 Data modeling

4 Metrics

5 Unusual data sources

6 Categorical data

7 Categorical data: Advanced methods

8 Time series data: Data preparation

9 Time series data: Analysis

10 Rapid prototyping: Data analysis

11 Rapid prototyping: Creating the proof of concept

12 Iterating on someone else’s work: Data preparation

13 Iterating on someone else’s work: Customer segmentation

A Python installation instructions