Time for our third strategy: good fact-checkers read “laterally”, across many connected sites instead of digging deep into the site at hand.

When you start to read a book, a journal article, or a physical newspaper in the “real world” you already know quite a bit about your source. You’ve subscribed to the newspaper, or picked it up from a newsstand because you’ve heard of it. You ordered the book from Amazon or purchased it from a local bookstore because it was a book you were interested in reading. You choose a journal article either because of the quality of the journal article or because someone whose expertise and background you know cited it. In other words, when you get to the document you need to evaluate, the process of getting there has already given you some initial bearings.

Compared to these intellectual journeys, web reading is a bit more like teleportation. Even after following a source upstream, you arrive at a page and site and author that are often all unknown to you. How do understand the author’s qualifications or the trustworthiness of the site?

Researchers have found that most people go about this the wrong way. When confronted with a new site, they poke around the site, and try to find out what the site says about itself, by going to the “about page”, clicking around in onsite author biographies, or scrolling up or down the page. This makes no sense. If the site is untrustworthy, then what the site says about itself is most likely untrustworthy as well. And even if the site is generally trustworthy, it is inclined to paint the most favorable picture of its expertise and credibility possible.

The solution to this is, in the words of Sam Wineburg’s Stanford research team, to “read laterally”. Lateral readers don’t spend time on the page or site until they’ve first gotten their bearings by looking at what other sites and resources say about the source at which they are looking.

For example, when presented with a new site that needs to be evaluated, professional fact-checkers don’t spend much time on the site itself. Instead they get off the page and see what other authorities have said about the site. They open up many tabs in their browser, piecing together different bits of information from across the web to get a better picture of this site where they’ve landed. Many of the questions they ask are the same as the vertical readers scrolling up and down the pages of the source they are evaluating. But unlike those readers, they realize that the truth is more likely to be found in the network of links to (and commentaries about) the site than in the site itself.

Only when they’ve gotten their bearings from the rest of the network do they re-engage with the content, with a better understanding as to whether to trust the facts and analysis presented to them.

You can tell a lateral reader at work: They have multiple tabs open, they perform web searches on the author of the piece and the ownership of the site. They look at pages linking to the site, not just pages coming from it.

When the lateral reader is looking for analysis, lateral reading helps the reader understand the perspective from which the site’s analyses will come. When the lateral reader is looking for facts, lateral reading helps the reader understand if the site has an editorial process or expert reputation that would allow one to accept a fact cited on the site as solid.

We’re going to deal with the second question here (factual reliability), while noting that lateral reading is just as important for the first question.