- Communications: misinformation and how to check it
How do fact-checkers evaluate a website?
Fact-checkers evaluate a website by lateral reading: instead of studying the site itself, they leave it within seconds, open new tabs and search for what other, independent sources say about who runs it and who funds it. A Stanford study found that this made professional fact-checkers faster and more accurate than historians and students, who were fooled by logos, domain names and polished pages.
The topic explained
Reading down or reading across
Most of us were taught to judge a website by looking closely at it: is it professional, does it have an About page, a .org address, references, an official-looking logo? Researchers call this vertical reading: moving up and down the page and asking the site about itself. The problem is that every one of those signs is cheap to fake. A slick design costs little, anyone can register a .org address, and an About page says whatever its owner wants.
Professional fact-checkers do something that looks lazy. They glance at the page, then leave it, open new tabs and search for the organization's name, its people and its money. This is lateral reading: reading across the web about the source, not down the source itself. They want to know what independent sources, such as news reports, encyclopedias and registers, say about who is behind the site, before they spend time on what the site says.
The Stanford study
Sarah McGrew and Sam Wineburg of the Stanford History Education Group compared three groups of experienced internet users: 10 historians with doctorates, 10 professional fact-checkers from news organizations and 25 Stanford undergraduates, 45 people in all. Their results appeared as a working paper in 2017 and in the journal Teachers College Record in 2019. One task asked them to judge information about bullying from two sites: the American Academy of Pediatrics, a large professional association of children's doctors, and the American College of Pediatricians, a small, socially conservative group with a similar name.
Many historians judged both sites credible, and students tended to trust the smaller group's site more. The fact-checkers quickly searched for the second group and learned what it was. Historians and students were swayed by logos, domain names and academic citations; the fact-checkers reached better-supported conclusions in a fraction of the time. Expertise in reading documents, the researchers concluded, does not transfer automatically to the web.
Click restraint and the SIFT moves
The fact-checkers also showed click restraint: after a search, they read the list of results, their snippets and addresses, before choosing where to click, instead of opening the first result. The educator Mike Caulfield turned these habits into four moves called SIFT: Stop and ask whether you know the source; Investigate the source; Find better coverage, other reports of the same claim; and Trace claims, quotes and pictures back to their original context. They are explained in Verified (2023), a book by Wineburg and Caulfield.
The moves are fast on purpose. Most claims online deserve thirty seconds, not thirty minutes, and the first move, stopping, is the one that prevents most mistakes.
A worked example and a warning
Imagine a made-up site, the “Center for Healthy Screens,” with a .org address, a clean design and many footnotes, claiming that a certain phone is the safest for teenagers. Vertically it looks excellent. Laterally, a search for its name finds a news report that it is funded by the company that makes that phone. The footnotes did not lie; the site simply was not the neutral source it seemed. One warning: lateral reading is not looking up a single source and trusting it blindly. An encyclopedia entry is a starting point; the goal is several independent sources that agree.
Practice: evaluating a website
For 12th grade communications students. Every site, organization and funder in the examples is made up. Answer from what is printed on the page, then check yourself against the answers.
Free, no sign-up. The answer key prints on the second page. Regenerating a new worksheet on the same topic needs a free account.
Vertical or lateral? (a) Reading the site's About page. (b) Searching for the organization's name in a news archive. (c) Checking whether the site has a .org address. (d) Looking up who funds the organization.
Fill in the missing tab in the diagram, and explain what it tells a lateral reader.
A vertical reader asks the site about itself; a lateral reader asks others. Read the chart. How many people took part, and what percentage of them were students?
45 participants. McGrew and Wineburg, published 2019. Why were historians, who are experts at judging documents, fooled more often than fact-checkers?
What is click restraint, and why does it help?
Fill in the missing SIFT move, and give an example of it.
Mike Caulfield's SIFT moves. A made-up site, the Center for Healthy Screens, has a .org address, footnotes and a clean design. A search shows it is funded by a phone maker. What should a reader conclude?
A classmate says: “Lateral reading means checking Wikipedia and believing it.” Correct the statement.
Common questions
What is the difference between lateral and vertical reading?
Vertical reading stays on a website and judges it by its own content and design. Lateral reading leaves the site and searches other, independent sources to learn who is behind it and whether its claims hold up.
Is a .org website always reliable?
No. Anyone can register a .org address, including advocacy groups and companies. The address says nothing about who funds a site or whether it is accurate; lateral reading does.
What does SIFT stand for?
Stop; Investigate the source; Find better coverage; Trace claims, quotes and media to their original context. It is a set of quick checking moves developed by Mike Caulfield.
Who studied how fact-checkers read websites?
Sarah McGrew and Sam Wineburg of the Stanford History Education Group, who compared 10 historians, 10 fact-checkers and 25 students in a study published as a working paper in 2017 and in Teachers College Record in 2019.
How long should checking a website take?
Often less than a minute. Fact-checkers in the Stanford study reached better conclusions in a fraction of the time the others took, because a quick search about the source answers the key question faster than reading the site.

Still have a question? Ask Lessi
Any question about this subject. Lessi answers right here, step by step.
Try asking
Asking needs a free account; your question is kept while you sign up. Answers are written by AI and can contain mistakes. Lessi doesn’t give medical or legal advice.
Learn it with a teacher
Learn this topic live with Lessi
Lessi, the AI teacher, explains it out loud on a board, with short quizzes, at your pace. The first 60 minutes are free.