Reading to write: a framework for learning the art of statistical storytelling

Deborah Nolan, Sara Stoudt

Significance
December 2, 2020

Introduction: We’ve cleaned and wrangled the data, we’ve made picture-perfect plots, and we’ve performed our statistical analysis. Now it’s time to share our findings with the world. This is a step that novices and experts alike balk at. That blank screen of our future manuscript is intimidating.

Instead of staring at a blank screen waiting for inspiration to hit, we can take action. In fact, even now, as you read this magazine, you can get ready to write. Learning how to write can start with reading data-related articles and examining how the author organises and writes about their findings. When we “read to write”, we identify the main components of a data analysis and notice how the author brings these components together to form a logical and compelling story. During this deliberate process, we discover examples and templates that we can use to organise our own work and write about our findings.

It can be daunting to develop an “ear” for writing. However, reading to write is a skill that can be honed through practice. To help you
get started, we have developed a three-step framework. We walk through each of these steps as we read “Nightingale’s overlooked Scutari statistics”, an article appearing on pages 29–33 of this issue of Signi cance.



Featured Fellows

Sara Stoudt

Statistics, UC Berkeley
Alumni - BIDS Data Science Fellow