What does an information designer with access to Google Trend data do?
Make a data viz about data viz. At least, that’s what the designer Anna Vital did. Her interactive graphic, called The Visualization Universe, uses 10,000 data points to show how different types of charts have varied in popularity over the past 12 months, based on how many people searched for that graphic on Google.
Among other things, it’s an intriguing look at data’s role in the 2016 election. In many of the most popular charts, there’s a noticeable bump in the trend line around the election–likely because many people were searching for answers through data.
The Visualization Universe is part of a bigger project from Google News Lab, which creates tools for journalists to use. Since the beginning of 2016, the lab has been asking designers to make visualizations using Google Trend data however they see fit. The only requirements are that their output is friendly on mobile, technologically innovative, and interesting.
Vital’s project is organized so that you can list different types of charts, books, and tools by name, search interest, and change in search interest–the last one reveals what is growing the most in popularity. A sparkline below each item shows its relative change over time in a simple line graph. That’s a particularly relevant area to delve into, because it shows how tastes in data viz are shifting.
“There’s been a massive boom in visualizing data. Now that boom is maturing, we’re a lot more informed,” says Simon Rogers, the data editor at the Google News Lab who previously worked at The Guardian doing data journalism. “We’re not overwhelmed by pretty pictures anymore. It’s very much about conveying info. We thought we’d looked at the ways people look at data using Google Trends as a proxy.”
The 2016 election, in particular, was something of a reckoning. Recall that almost no data analysts predicted it. That was partly a data problem and partly a visualization problem. Many data designers put together cutesy, overly clever visualizations that skewed the narrative or simply did not allow for more open-ended interpretation. As my colleague Mark Wilson wrote last November, “The best data publications in the world ultimately catered to our need for simplified narratives–to the point that even lauded data designers got lost inside it all.”
So it’s perhaps not surprising that Vital’s visualizations show interest in fairly straightforward charts growing over time–particularly cartograms, timelines, treemaps, and word clouds. These are difficult visualizations to misinterpret. There’s no definitive way to tell if that was due to the election, but Rogers acknowledges that the election “undoubtedly led to a high interest in all types of data visualization.”
The rise in the use of cartograms–abstract maps of statistical information–especially points to how data viz designers were trying to understand the nuances country’s politics. “It’s not just a state by state map,” he says. “You can’t show that anymore. You want to show things in a more representative way.”
Overall, the Visualization Universe reveals that bar charts, Gantt charts, and histograms are the most-searched-for types of charts. There are also sections that assess the popularity of data viz books and visualization tools (Edward Tufte‘s The Visual Display of Quantitative Information was the most searched-for book, and Excel and PowerPoint were the top general tools). The Visualization Universe will be updated with new data every day, so it’s truly tracking the evolution of how people are searching for data live.
“What the search data gives you is the level of honest interest,” Rogers says. “It takes you beyond the echo chamber of social media and you’re seeing what people really care about.”