Communicating With Data: The Art of Writing for Data Science

By Deborah Nolan and Sara Stoudt

Communication is a critical yet often overlooked part of data science. “Communicating with Data” aims to help students and researchers write about their insights in a way that is both compelling and faithful to the data. General advice on science writing is also provided, including how to distil findings into a story and organise and revise the story, and how to write clearly, concisely, and precisely. This is a useful resource for students who want to learn how to write about scientific findings, and for instructors who are teaching a science course in communication or a course with a writing component.

The book consists of five parts: Part I helps the novice learn to write by reading the work of others. Part II delves into the specifics of how to describe data at a level appropriate for publication, create informative and effective visualisations, and communicate an analysis pipeline through well-written, reproducible code. Part III demonstrates how to reduce a data analysis to a compelling story and organise and write the first draft of a technical paper. Part IV addresses revision; this includes advice on writing about statistical findings in a clear and accurate way, general writing advice, and strategies for proof reading and revising. Part V offers advice about communication strategies beyond the page, which include giving talks, building a professional network, and participating in online communities. This book also provides 22 portfolio prompts that extend the guidance and examples in the earlier parts of the book and help writers build their portfolio of data communication.

The book provides concrete models of strong and weak writing, as well as a variety of writing activities, including more freeform portfolio prompts. It also covers writing code and creating visualisations as important communication skills.

Table of contents

Reading Science Articles
Reading Materials Written for Broader Publics
Describing Data
Communicating Through Statistical Graphs
Communicating through Code
Organizing the Story
Writing the First Draft
Taking Care with Statistical Terms
Crafting Words and Sentences
Revising: Drafts #2 Through…
Embracing Your Role as a Scientist
Building a Portfolio

Deborah Nolan is a professor of Statistics at the University of California, and Sara Stoudt is a lecturer in Statistical & Data Sciences at Smith College.

More information