this space intentionally left blank

January 17, 2024

Filed under: journalism»data

Add It Up

A common misconception by my coworkers and other journalists is that people like me — data journalists, who help aggregate accountability metrics, find trends, and visualize the results — are good at math. I can't speak for everyone, but I'm not. My math background taps out around mid-level algebra. I disliked Calculus and loathed Geometry in high school. I took one math class in college, my senior year, when I found out I hadn't satisfied my degree requirements after all.

I do work with numbers a lot, or more specifically, I make computers work with numbers for me, which I suspect is where the confusion starts. Most journalists don't really distinguish between the two, thanks in part to the frustrating stereotype that being good at words means you have to be bad at math. Personally, I think the split is overrated: if you can go to dinner and split a check between five people, you can do the numerical part of my job.

(I do know journalists who can't split a check, but they're relatively few and far between.)

I've been thinking lately about ways to teach basic newsroom numeracy, or at least encourage people to think of their abilities more charitably. Certainly one perennial option is to do trainings on common topics: percentages versus percentage points, averages versus medians, or risk ratios. In my experience, this helps lay the groundwork for conversations about what we can and can't say, but it doesn't tend to inspire a lot of enthusiasm for the craft.

The thing is, I'm not good at math, but I do actually enjoy that part of my job. It's an interesting puzzle, it generally provides a finite challenge (as opposed to a story that you can edit and re-edit forever), and I regularly find ways to make the process better or faster, so I feel a sense of growth. I sometimes wonder if I can find equivalents for journalists, so that instead of being afraid of math, they might actually anticipate it a little bit.

Unfortunately, my particular inroads are unlikely to work very well for other people. Take trigonometry, for example: in A Mathematician's Lament, teacher Paul Lockhart describes trig as "two weeks of content [...] stretched to semester length," and he's not entirely wrong. But it had one thing going for it when I learned about sine and cosine, which was that they're foundational to projecting a unit vector through space — exactly what you need if you're trying to write a Wolf3D clone on your TI-82 during class.

Or take pixel shader art, which has captivated me for years. Writing code from The Book of Shaders inverts the way we normally think about math. Instead of solving a problem once with a single set of inputs, you're defining an equation that — across millions of input variations — will somehow resolve into art. I love this, but imagine pointing a reporter at Inigo Quilez's very cool "Painting a Character with Maths." It's impressive, and fun to watch, and utterly intimidating.

(One fun thing is to look at Quilez's channel and find that he's also got a video on "painting in Google Sheets." This is funny to me, because I find that working in spreadsheet and shaders both tend to use the same mental muscles.)

What these challenges have in common is that they appeal directly to my strengths as a thinker: they're largely spatial challenges, or can be visualized in a straightforward way. Indeed, the math that I have the most trouble with is when it becomes abstract and conceptual, like imaginary numbers or statistical significance. Since I'm a professional data visualization expert, this ends up mostly working out well for me. But is there a way to think about math that would have the same kinds of resonance for verbal thinkers?

So that's the challenge I'm percolating on now, although I'm not optimistic: the research I have been able to do indicates that math aptitude is tied pretty closely to spatial imagination. But obviously I'm not the only person in history to ask this question, and I'm hopeful that it can be possible to find scenarios (even if only on a personal level) that can either relate math concepts to verbal brains, or get them to start thinking of the problems in a visual way.

Past - Present