Jen Christiansen is senior graphics editor at Scientific American and has written the book Building Science Graphics: An Illustrated Guide to Communicating Science with Diagrams and Visualizations, due to be released in December 2022 (AK Peters Visualization Series, CRC Press). I had a chance to chat with her about the book in particular and science communication in general. Here is the edited version of our conversation:
Núria Altimir: I have not seen the book beyond the cover, but I see it includes the word BUILDING and the illustration on the side implies pieces clicking together, plus it is described as a guide. Are you giving in the book the building blocks for making a good science graphic?
Jen Christiansen: Yes, and no. I wanted it to be a practical guide, so that somebody who is not necessarily trained as a designer—like a scientist who needs to create a graphic—could walk through a process that would allow them to create one. The book contains design principles and science communication principles: Those ideas could be thought of as building blocks. But the most practical content resides in two chapters that are comprised almost entirely of large flowcharts. Those flowcharts walk you through the steps of creating a graphic. For example, remembering to check in with collaborators, how to do research, that sort of thing. So, it’s a process-centric book, but it doesn’t necessarily give you prefabricated solutions that you’re supposed to click together.
NA: To the GRAPHICS in the title—what kind of graphics does the book focus on?
JC: It’s focused on information graphics. I think of information graphics as a continuum, with figurative drawings on one end and abstract representations on the other. By “figurative” I mean objects or scenes that are drawn—more or less realistically. On the abstract end is where I think data visualization comes in. You’re taking quantitative information about objects or processes, and abstracting that information into a colour or a tone or a position within a frame. My book focuses mostly on the middle portion of the continuum—an area that holds explanatory diagrams. Sometimes graphics in that zone may hold some figurative elements AND some dataviz elements. But it’s sort of in this weird space that I think is lacking in resources right now. The data visualization community has really matured, and there are lots of great data visualization books available right now. But it feels like we’re missing something in the explanatory illustrated diagram realm. Especially now that the Malofiej Infographics Summit has been paused (a conference, workshop and awards program hosted for many years by the Spanish chapter of the Society for News Design). The Malofiej community provided a great place to connect and discuss explanatory illustrated diagrams. This book is, in part, to help fill that gap. And I’d love for it to also keep dataviz practitioners from losing sight of what happens at the edge of their discipline.
NA: Did you miss having a similar resource back in your studying days?
JC: In college, I double majored in geology and studio art, and then I studied scientific illustration in graduate school. Most of my studio art resources at that time focused on technique. I don’t recall many textbooks. Except for some of Edward Tufte’s books, and the Guild Handbook of Scientific Illustration. But none of those really brought together science communication and information design in a comprehensive yet concise and practical way. Don’t get me wrong. All of those books influenced me a great deal. And I still refer to them today. But I was missing an engaging resource that included opinion rooted in experience, and lots of practical advice, and further recommended readings.
NA: The book has, of course, a lot of illustrations. Where have you found the material?
JC: Some are my own work. Especially the sketches. But many are from other people. Some are from folks that I collaborated with—which allowed me to discuss the project from the viewpoint of someone actively involved in shaping design decisions for a particular project. Others are final graphics from other designers that I think exemplify a particular design concept particularly well. There are also several “spotlight” sections in the book. These pages feature the graphics-building process from the point of view of several different designers, and feature projects from the sketch stage all the way through to final art.
NA: How much of this book is Jen? Is it you wanting to explain your experience to others? Or is it a review of what everybody knows about creating science graphics?
JC: It’s both. The book definitely includes information rooted in my experiences. But I used writing it as an excuse to carve out time to read books and papers, and spend time thinking more thoroughly about how my experiences fit within the larger world of science graphics. And also acknowledging that there are areas that I’m not an expert in, but need to be covered in a book like this. For example, I’m not an expert in accessibility. But I couldn’t write a book about graphics without addressing it. In that section, I rely heavily on the expertise of others. I decided to write this book now because it was an opportunity to have Alberto Cairo as an editor. His book The Functional Art remains one of my favorites, largely because he pairs his personal experiences with a broader overview. I wanted to take the same approach to Building Science Graphics, so having him as a mentor was an opportunity I didn’t want to miss out on.
NA: In a previous conversation of yours with a visual perception researcher, you said you were interested in knowing “how people read the graphics”. After all the digging done while writing the book, have you managed to find out?
JC: It’s frustrating as a practitioner! We want evidence to inform decisions on how to build graphics. But a lot of the research studies are—by necessity—focused on really specific variables. You can’t just look up the answer to support every design decision: Researchers aren’t designing studies that replicate every variable for the graphic that I’m personally building. So yes, there’s a lot of research cited in this book. But it also includes caveats. Along with tips from folks including Sheila Pontis on how to dig in and start collecting data about how our own graphics are received by our own audiences.
NA: Let’s take the other big word on your book cover: SCIENCE. What would you say is particular to science communication, such that we need a book focused on it?
JC: Science graphics and science communication fill this weird space in which we’re putting out information and expecting people to take that evidence in, and have it inform their decisions in a rational manner. But not everyone fundamentally understands how the practice of science is conducted. Science is a self-correcting enterprise. Interpretations shift and build as the evidence builds. Different people are studying different parts of a whole. It’s a process that involves both replicating and challenging each other’s work. And over time, you get a greater understanding of something. “Science” isn’t going to land on a single correct answer right away in all cases. So how does that inform your graphic? I think it’s important to nod to the process somehow. Provide context. Be wary of absolutes. Embrace uncertainty and complexity, and avoid distilling things down into pieces that can be easily dismissed as our understanding of a phenomenon evolves.
NA: I have a science background myself, and I’ve always wondered where the science finishes and the “normal topics” start. Wouldn’t we like to see all kinds of things explained as rigorously as you would explain so-called scientific contents? JC: Yeah, once you start pulling on that thread, it shakes your idea of what’s certain and what’s not. To my mind, it all comes down to the goal of the graphic. For example—on the decision to include nods to uncertainty—if the graphic’s goal is to specifically be a tool to aid decision makers, then yes, be transparent about the fuzzy edges. Whether it’s rooted in a “science” topic or not. But what if it’s something like a diagram about describing photosynthesis? Hypothetically, let’s say that most of the process is known very well. But there are a few details that remain a little fuzzy. Do you need to dramatically point out every piece of the puzzle that’s not completely known? I don’t think so. The same goes for topics not rooted in science. If the goal is to provide a peek at the overall gist, then I don’t think that every caveat needs to be addressed visually.
NA: More about boundaries: Storytelling and attention-driving techniques are used to prime and guide the reader through the information, especially when the information is complex. How do you do that so that the reader does not have the feeling of being misguided, manipulated, or played along some agenda? Is there a blurred edge in this?
JC: To my mind, in dataviz, but also other kinds of visualization, there are two purposes: analysis or communication. In analysis mode, you don’t necessarily want to have a preconceived story. You want to be looking at the topic with fresh eyes, and be open to seeing patterns that you might not be specifically looking for. For illustrated explanatory graphics, that’s the research or reporting and exploratory sketch stage. But by the time you get down to creating a visualization to communicate something to an audience, whether it be a chart or drawing, you need to have a well-formulated concept of what it is that you’re trying to communicate. You may call it the story, or the point of the graphic, or the goal of the graphic. Whatever you call it, that goal needs to be crystal clear in the designer’s mind. And then design decisions need to be made to support that goal. Especially in the world of journalism. We’re generally not in the business of creating visualizations for somebody to explore on their own and find their own story. People want to know what the experts say. I don’t think that’s a form of manipulation. It’s asking questions of experts or data and then reporting back to your audience what you’ve learned. I think it’s great when the supporting context remains, but doesn’t overwhelm the primary storyline. That’s what I love about graphics. You can create levels of information within a single frame. The full dataset may remain intact in a somewhat transparent gray, for example, but the points that help guide the story may be in a darker color, or highlighted with annotations. I think it gets tricky when people start to intentionally mislead or mischaracterize, or craft a story that isn’t actually supported by the information being shown. But I feel like that’s a different issue.
NA: I brought this up because initially thought graphs ought to be objective and neutral. But with urgent subjects—like the Covid pandemic, or climate change—I find it difficult to stay neutral. Should we not portray a sense of danger and risk in all graphs on climate change? Or is that showing too much of an agenda?
JC: The idea that an objective and neutral viewpoint is even possible is the crux, right? Lewis Raven Wallace’s The View from Somewhere is a great resource on this topic as it relates to journalism. They credit their collaborator Ramona Martinez with saying “objectivity is the ideology of the status quo.” I think that’s also being reflected by some of the movements within the data visualization community, like data humanism and data feminism.
NA: Who do you think will benefit most from your book? The journalists, the people working in data visualization, or the actual researchers?
JC: The primary audience is research scientists and students. And science communication professionals. That said, it should be really useful for anyone that needs to communicate scientific content with visuals.
NA: Because they often neglect the communications part. Some researchers are naturally talented at communications, but many just ignore it. They have internalised one way of making graphs that are understandable only by their peers . To these people we are recommending the book , right?
JC: Yes, in part because it also prompts people to think about who they’re creating the graphic for. I think designers and journalists have that baked in a little bit more to our process. But scientists are often more focused on the content. Not the audience or outlet.
NA: To redo your visuals for different uses is a lot of work. Is it really the researcher who should be doing that, or does the role of science communicator need to be more prominent in research teams, because that’s the person who understands both sides and can really focus on creating good graphics? Perhaps science communicators or visual artists or something need more promotion inside academia?
JC: Excellent point. There’s a chapter in the book about collaborations. You may have a vision or a need, but not the time or the expertise to execute the graphic. As a scientist, it’s really useful to simply know how to communicate with a designer, and as a designer it’s really useful to know how to communicate with a scientist. This includes understanding contracts, pricing, image rights, et cetera.
NA: Yeah, I was thinking more like somebody who is in the group. Like you have a lab assistant, you have a visual assistant.
JC: That would be awesome. If principal investigators have the budget to do so I think it can definitely be a bonus.
NA: The graphs that scientists generate are often style-bound to what the scientific journals demand. How far away are we from seeing the scientific journals asking for graphs that are more readable to anyone?
JC: Yeah, that’s a great question. I don’t work at a journal, but there seem to be a lot of production-related things in play that limit what types—and sizes—of visuals can be submitted. My impression is that the supplementary material documents are starting to be more open to things like interactives and videos. I also have noticed more and more publishers providing trainings or services to help scientists become better writers and communicators. But we should realize that some visual styles used in scientific journals are part of an established language. It’s visual jargon. Jargon can be really efficient when you’re talking with other people who are fluent in that language. It’s the same thing with visuals. I don’t want it to go away for the people who need to be able to speak to each other in that really specialized language. Great, you do that for the primary literature. But eventually it should be translated into a visual language that more people can understand.
NA: Outside the academy: how should we treat our so-called general audience? It is often assumed that people know nothing, and they expect to be entertained, that things need to be simplified for them. Are we not losing a chance for them to learn something?
JC: I’m a huge proponent of retaining as much complexity as possible, and providing the audience with the tools they need to read it. Nigel Holmes, Alberto Cairo and other designers have talked about: “Clarify, don’t simplify.” I try to use that as a mantra. What can I do as a designer to help somebody understand what they’re looking at without distilling it down into something that’s just too simple: things like annotations, maybe a playful style that draws somebody in. It’s about knocking down barriers of entry. There are different strategies to do that. You can provide an introductory graphic that gets people up to speed who might not have a background in a particular topic, and then show the latest cool discovery. That’s really useful in news stories on science topics. If people understand that background already, they can just buzz past the intro graphic and get to the new stuff. By separating the material out into chunks, you provide more options for people at different levels of familiarity with the topic to get involved.
NA: And since this interview is for DVS, what do you think someone focused on data visualization will extract from your book?
JC: I think the main thing that a data visualizer will extract from the book is that creating an explanatory graphic—like creating a chart—is a process. Also, as communities become more and more specialized and insular, I fear that we stop learning from disciplines that are on the periphery of our specific niche. Often data visualizations don’t live in isolation. They work in tandem with other visual explanatory bits. I’m hoping that this book will help data-centric folks bridge some of the gaps, and help them think through how to use data visualization and schematics together.
Núria Altimir is a natural science researcher turned illustrator, works as a freelancer science illustrator and data visualizer. She has a Biology degree, a PhD in Forest Sciences and more than 15 years of experience in academic research. During the past 5 years she worked as graphical officer supporting researchers at the University of Helsinki. She is Barcelona born and long-time Finland resident, trying to get her analytical and creative selves to get along.