SQL for Data Scientists

Introduction

Who I Am and Why I'm Writing About This Topic

When I was first brainstorming topics for this book, I used two questions to narrow down my list: “Who is my audience?” and “What topic do I know well enough to write a book that would be worth publishing for that audience?”

The first question had an easy initial answer: I already have an audience of data-science-learning Twitter followers with whom I share resources and advice on “Becoming a Data Scientist” that I could keep in mind while narrowing down the topics.

So then I was left to figure out what I know that I could teach to people who want to become data scientists.

I have been designing and querying relational databases professionally for about 17 years: first as a database and web developer, then as a data analyst, and for the last 5 years, as a data scientist. SQL (Structured Query Language) has been a key tool for me throughout—whether I was working with MS Access, MS SQL Server, MySQL, Oracle, or Redshift databases, and whether I was summarizing data into reporting views in a data mart, extracting data to use in a data visualization tool like Tableau, or preparing a dataset for a machine learning project.

Since SQL is a tool I have used throughout my career, and because creating and retrieving datasets for analysis has been such an integral part of my job as a data scientist, I was surprised to learn that some data scientists don't know SQL or don't regularly write SQL code. But in an informal Twitter poll ...