Data doesn’t have all the answers: Amanda Shendruk on balancing feelings with facts
Plus: How has your job changed your personality?
Welcome to The Writethrough. Last week journalist and media consultant P. Kim Bui wrote a guest essay on the diminishing appeal of leadership roles in journalism. This week: Visual journalist Amanda Shendruk on using data to understand the world.
How has your job changed your personality?
I used to fight with my dad a lot more than I do now. Part of the change is maturity on my part, I’m sure, and the mellowing that comes with age on his. But I also think it’s because being a journalist for so many years has changed the way I behave in conversations with people whose views differ from my own.
When you’re a reporter, you can’t get upset when people say things you disagree with, even if a source makes a comment that really pushes your buttons. I want people to tell me what they really think, so I’ve learned to keep my cool. I ask questions — both because I’m curious and because that’s a big part of my job. I’ve noticed sometimes it’s the act of asking questions that actually awakens my genuine curiosity. I didn’t actively make a decision to start acting differently with my dad; I just started going about conversations with him on tricky subjects with the same inquisitive mindset I might have with a source. Maybe there are drawbacks here – I used to get annoyed when my mom, a retired psychologist, would put on her therapist hat with me — but overall I think we’re both happier as a result.
Of course all journalists are still human beings with their own inevitable biases, and there’s a valuable, ongoing conversation in the media industry about the perils of believing too much in your own objectivity. But I’ve been thinking a lot recently about how our jobs, and the ethics that form the bedrock of our professions, reshape our personalities and even influence the way we act in our personal lives. I’m sure this is true for all kinds of roles. So whatever your job, whether you’re a journalist, nurse, teacher, real estate agent, or something else altogether, I’d love to hear from you about how it’s informed the person you are now. (Potentially for another, longer post!) Drop me a line at sarahlizchar [at] gmail.com, or leave a note in the comments. — Sarah Todd
Two can’t-miss workshops on navigating layoffs and the art of reaching out
A few weeks ago, we shared advice from media professionals on the art of networking and most importantly, how to do it on your own terms. One great person to hear from on this topic is Rosie Spinks, our former colleague from Quartz, who is writing a book on building a village in a culture and society that often encourages putting up more walls than any form of interdependence. On Thursday, Feb. 19, Rosie will be hosting a subscriber-only workshop called “The art of reaching out,” based on her experience as the author of a forthcoming book and journalist.
Meanwhile, career coach Phoebe Gavin (also a former Quartz colleague and a frequent contributor to The Writethrough!) is offering a free workshop for recently laid-off journalists next week, on Wednesday, Feb. 25. The community session promises to help people hit by cuts at Washington Post and elsewhere come up with a strategy to get them where they want to go next.
We highly recommend both!
Why data is often controversial: An interview with visual journalist Amanda Shendruk
Amanda Shendruk is the journalist behind Not-Ship, “a data-focused, visually ambitious newsletter trying to navigate the shifts and currents of our messy world.” She’s reported for the Washington Post, Quartz, The Guardian, and Maclean's Magazine.
How do you feel visual journalism has changed since you began your career, say over the last 10–15 years?
When I talk about visual journalism, I mean the kind of work I do: interactive graphics, data visualizations, and big, built-out experiences.
When I started out, that kind of work barely existed as a defined field. There were a few teams—like at The New York Times—and a handful of books. “Snow Fall” at the Times really changed how people thought about what was possible, and I came in just as that wave was building, maybe 15 years ago.
I was in Canada then, and there wasn’t a single newsroom hiring for this kind of role. Now there are at least a few people doing this work at a few outlets, so it has grown a lot. At the same time, I worry we may have hit “peak visualization” in newsrooms. We’ve seen big layoffs, including at places like The Washington Post’s graphics desk—roles that once felt relatively “safe” because they were seen as the future of journalism.
These jobs are highly specialized and hard to hire for, so competition is weirdly shaped: there aren’t many positions, but there also aren’t that many people with the skillset. The real issue feels less about demand for the work and more about the overall instability of the industry and the higher cost of hiring interactive journalists who can code, compared with, say, a couple of junior reporters.
How much of that do you think has to do with people’s attention spans? How do readers interact with visual journalism now?
At places like the Post and at Quartz before that, our visual and interactive pieces consistently did really well. It’s very clear to me that readers enjoy interactive experiences and stories that use data, visualized clearly, to explain something.
So I don’t think the pullback is because audiences don’t want this kind of storytelling. I think it’s about economics and business models. Interactive journalists tend to be more expensive because you’re paying for both journalism and technical skills. If a newsroom is under pressure, it’s easy to look at a role like that and think, “We could hire two junior writers for the same cost.”
In other words, I don’t see a format problem; I see an industry problem. The audience appetite is there. The business model isn’t.
Let’s talk about your newsletter. What are you trying to achieve with it, and how is it going so far?
It’s called “Not Ship”, and I launched it in October. The idea is to use data, data visualization, and explanatory graphics to help people make sense of an increasingly chaotic and uncertain world.
The newsletter came out of a period where I was dealing with a lot of change in my own career, while everything in the wider world also felt very unstable. I had many friends who were feeling that same uncertainty. I can’t personally solve the big structural problems behind that, but I do know how to work with data, and I use data to try to understand the world a bit better.
So the newsletter is my attempt to use data to “anchor” us—to give people something more solid than just vibes to hold onto. I want it to be a place where we can use data as a tool to think more clearly about what’s happening, without pretending data alone is enough.
Importantly, I don’t want the newsletter to be me just talking at people. The goal is to build a space where readers can respond, question, and discuss the data with me. I want it to feel like a conversation.
One example is your piece on American exceptionalism, which you revisited and which sparked a lot of engagement. Could you talk about that experience and what it showed you about your readers?
Over my career, I’ve spent a lot of time looking at international comparative data. Again and again, I’d notice the U.S. popping out as an outlier in all sorts of charts. Over time I started jotting those examples down whenever the U.S. looked markedly different from its peers.
I wanted to play on the idea of “American exceptionalism” in two senses: the political concept that America is somehow special, and the statistical sense of an exception—an outlier. I assumed I’d find a mix of positive and negative outliers. In reality, the data skewed heavily negative: things like poor life expectancy relative to health-care spending, extremely high incarceration rates, and so on.
I gathered those examples into a piece that argued: yes, America is often exceptional—but not in the flattering way the phrase usually implies.
The response was surprisingly constructive. Most readers engaged seriously with the data. I did get a few negative or hostile comments, but far fewer than I might have expected for a topic like that. The piece also spread beyond my usual subscriber base, including on Reddit.
For me, it was proof of concept that the newsletter could be more than just me publishing essays. It showed that if I put thoughtful, well-explained data work into the world, people would respond thoughtfully and want to have a real conversation about it.
That leads to a bigger question: why is data so often controversial?
I think a lot of people who don’t work with data regularly see it as inherently authoritative. A chart in a story can feel a bit like a doctor in a white coat: its mere presence signals, “This is objective and correct.”
But anyone who works with data knows that’s not quite true. Data can be framed and sliced in many ways—often without any malicious intent—that lead to very different interpretations. Change the time window and the trend can flip; change the comparison group and a pattern can disappear or emerge.
That doesn’t mean “anything goes,” but it does mean data isn’t as perfectly impartial as people sometimes imagine. When people realize that, it can be destabilizing. It’s also dangerous when we start treating data as if it has all the answers, or as if it’s somehow above questioning.
In the newsletter, I try hard to be transparent about my sources, methods, and the limits of the data. I’ll often point out where the data might be weak or incomplete, or where there are important questions it can’t answer.
A recent piece on “third places” is a good example. There’s a widespread narrative that third places—cafés, parks, community spaces—are in steep decline. But when I went looking for data, the best pre-COVID data I could find didn’t really show that decline. That was actually tricky to write, because the numbers didn’t line up neatly with what “everyone” seemed to feel was true. And we still don’t have good post-COVID data.
So, I use data to push back against pure intuition, but I also take feelings seriously. In that third-places piece, I stop halfway through and say, “Okay, the data doesn’t clearly support this decline narrative, but many of us *feel* like it’s true—so what might be going on?” Then I explore possible explanations with whatever evidence we have.
Data is a powerful tool, but it’s not the only thing we should consider. It should inform our thinking, along with our judgment, values and experience. But overall, I don’t think data is part of the conversation often enough.
You’re largely self-taught. Do you think people can still teach themselves data journalism now, or was that a particular moment in time?
In university, I worked at the student newspaper and got interested in the data side of things. I hunted down every book I could find on data visualization—there weren’t many then, unlike now—and I learned a lot from that early literature.
I also taught myself to code, which was the hardest part. I’m not great at sitting down and doing a full, structured course just for its own sake. I learn best when I have a concrete project and a deadline, which meant I was often “building the bus while driving it,” learning enough JavaScript or whatever I needed to make the thing work on the fly, while reading a lot of Stack Overflow.
It was doable, but pretty painful at times.
I actually think it’s easier to self-teach now than when I was starting out. There are many more books, courses, tutorials, and communities. And large language models like ChatGPT and Claude have changed the game. I use Claude to help with coding now; projects that might have taken me a week can sometimes be done in a couple of days because I can ask it to explain or debug things quickly.
So yes, I absolutely think people can still self-teach data journalism—and in many ways, the environment is far more supportive of that path than it used to be.
Given everything we’ve talked about—industry upheaval, AI, new tools—what do you see as the future of visual journalism?
AI is already changing my work. I use tools like Claude to help with coding. The work is still mine, but certain tasks are dramatically faster. That alone will reshape how visual journalists work.
But I don’t think the core question about the future of visual journalism is really about technology. It’s about whether journalism as an industry can find sustainable business models.
If the industry stabilizes, I think visual journalism will continue to thrive because it’s such a powerful way to communicate. If the industry continues to hollow out, visual journalism will suffer along with everything else. I don’t think it’s uniquely doomed or uniquely safe.
On a personal level, the newsletter is my attempt to carve out one possible future for myself in this landscape. I would have loved to do this kind of work at a big publication indefinitely, with a steady paycheck. But those institutions—places like The Washington Post—aren’t structured to support the kind of direct, two-way relationship with readers that newsletters offer.
We’ve seen how much people respond to more personal, less mediated journalism: podcasts, independent newsletters, online news creators. To me, that’s a signal that there’s real value in being able to speak directly to an audience and build trust with them over time.
That won’t, on its own, solve the business-model crisis for the whole industry. But for me, it’s one way of experimenting with what a viable future might look like.