You are probably no closer to finishing your next data project after reading this book.
I am painfully aware that the theory in this book is somewhat removed from the daily work of data journalism. You’re going to need practical skills like working with spreadsheets, cleaning data, coding up visualizations, and asking civil servants for explanations. I’ve covered none of this craft.
Yet all of this work is guided by old and deep principles. Journalists are latecomers to quantitative thinking. That’s unfortunate, because numbers can bring us closer to the truth. But only sometimes. Hopefully you now have a better sense of the limitations of data, and the ways we analyze and communicate data.
There’s a lot more to learn.
There are an endless number of technical concepts relevant to data work. I’ve tried to give an authentic taste of the state of the art, and Bayesian statistics and cognitive biases are at the forefront of contemporary practice across many fields. Still, these presentations do not have the depth and detail needed to do real work; no one is going to learn to do statistical analysis from what I’ve written. Not exactly.
The good news is you don’t have to learn everything at once. An education in statistics will give you powerful fundamentals that can be used to reason about subtle problems, but you won’t need to do that every day. Also, that’s what collaborators and mentors are for. A journalist’s primary responsibility is to the story, and technical mastery comes from the experience of many solved problems.
It’s not knowing everything that makes a technical professional, but being willing to find out. I’ve used standard mathematical language in an effort to help you find more information; with a search engine, knowing the true name of something gives you the ability to summon it at will. So don’t be surprised when you don’t know something. If you’re anything like me you’ll get the code wrong the first time, even when you do know what you’re doing. But never doubt that there is a logic underlying every equation and every line of code. These things are not magic; though the symbolic languages of data can be intimidating, there is nothing occult here.
My advice is to look always for the underlying sense of the thing, the plain-language explanation. This sense can be hard to find. When you ask a question like “why does a survey have a bell-shaped error distribution?” you will soon find yourself lost in inscrutable proofs, answers that seem to presuppose you already know, explanations that don’t really explain. This is an unfortunate comment on the sate of our educational materials, but don’t lose hope! Keep searching until you find an answer that makes sense.
Yet a technician is not a journalist. What will you be able to do with all of this understanding and ability?
Like any medium, it can take a while to find your voice in data journalism. Sure, you can do analysis and visualization and all the rest of it—but what are you saying? What questions are you asking? What is it that is so important, so urgent, that you must command a stranger’s time to tell it to them?
I don’t know of any way to discover what you want to say other than saying it. Just write. And report and code and visualize, but whatever else you do, put your work into the world. Then do the next one. As Steve Jobs said, real artists ship.
If you continue your study of the deep workings of data, you will discover entire worlds. You will retrace thousands of years of inspired ideas, re-experiencing each little epiphany as your own. You will gradually arrive at one of the most exciting frontiers of human thought, and you will join professionals in many other fields who are transforming their work through data. Quantitative ideas now pervade every aspect of the functioning of society, from health to finance to politics. It’s impossible to understand the modern world without understanding data.
And if you do understand data, you will begin to see stories that others literally cannot imagine. We need those stories told. That is, perhaps, the best possible argument for learning more.