Some thoughts after musing about all the new types of social science methods and financial analysis that journalists should know and understand to function in a new world of data: If it’s this hard for us, what about for readers? How hard do they have to work to figure out what we mean when we present them with a complex dataset or argument?
To be sure, it is the job of journalists to render complex ideas into something an average person can understand; if they don’t get it, we haven’t done our job.
And it’s often not necessary for someone to grasp all the details of some idea or method to be able to get the gist of a story – so while a journalist ought to know what the exact meaning of “margin of error” is when he or she writes a story about a survey, the reader only needs to have a general idea what the term means.
In the case of The Wall Street Journal’s Pulitzer-prize winning series on options backdating, the paper took a pretty complicated algorithm that looked at how unlikely some events were and simplified significantly. And it worked.
A Wall Street Journal analysis suggests the odds of this happening by chance are extraordinarily remote — around one in 300 billion. The odds of winning the multistate Powerball lottery with a $1 ticket are one in 146 million.
But if we were putting up an entire database of companies and the details of when their options were dated, how well could we have explained or visualized the math behind that calculation? Would it have mattered? Or do we tell readers to take our word for it?
That’s where it gets complicated. Narrative is one thing – we write a tale, and readers follow along. We fill in with numbers and other backup for our thesis as we go along, and if it convinces readers, it’s done its job.
What about when we present data and other information online? How do we make sure people know what we’re presenting as they explore it? In the case of the Obameter at Politifact, it’s pretty clear:
PolitiFact has compiled more than 500 promises that Barack Obama made during the campaign and is tracking their progress on our Obameter.
It doesn’t say it compiles every promise Obama has made, and it’s pretty clear that the ratings are subjective ones, based on reporting. You know what you’re getting pretty quickly. That’s the appeal of a very simple and straightforward interface.
In the design of WhoRunsHK, which shows relationships among key people in Hong Kong, which we tried to keep as simple and intuitive as possible, you still have to take some time playing with it to get the most out of the system. Or at least watch the video.
The Washington Post’s Top Secret America project has fantastic information behind it, but again you have to invest a fair amount of time into learning how to master that interface before you really reap rewards from it. Similarly China Vitae isn’t self-explanatory, and neither is Open Secrets.
None of this is bad – in fact, this is good, because it shows experimentation and innovation. But that doesn’t mean it’s easy for readers. How much time is any reader willing to invest before they decide it’s too much work?
Part of the issue is that we’re at the very beginning of a new medium of communication, so we’re still making up the rules as we go along. Information design and interface design is still relatively new, at least in terms of what we’re trying to communicate these days – and these aren’t areas that journalists, by and large, have any great expertise in. But we’ll need to learn, fairly quickly.
And over time, we may well have more and more standardized formats for displaying information, the way the proliferation of Microsoft Office standards have made it easy for everyone to understand the concepts of cut and paste and drag and drop. As we build out more and more databases for readers to explore, we should be also getting used to standard formats – so that readers can adjust easily from one site to another, and spend more time getting information than learning how to get information.