Here are some things students sometimes believe:

• They believe that .org's are more trustworthy than .com's.
• They believe that a good looking layout and few typos mean something is trustworthy.
• They think pages that have ads on them are less trustworthy than those that don't.
• They think that footnotes automatically make a site credible.

Now we are taking a new approach to evaluating sources. Before you learn to read the web you may have to unlearn some of the stuff you've been taught!

Sometimes they weren't checklists, but a set of "things to look for in a web page."

What the approaches had in common was this: you'd come to a webpage and look for signals that the page was trustworthy on the page itself. You'd ask questions like

• Does this look professional?
• Are there spelling errors?
• Is there scientific language?
• Does it cite sources?

The "checklist" idea? Students were told the more "good" things the page had the more "trustworthy" it was.

Except there are three problems with this approach. Big problems.

Problem 1: The signals are meaningless

Picture of Word document with autocorrect

Most things people are told to look for? They're easy to fake. It's really cheap to get a good-looking site nowadays. You can be a disinformation agent from China or Russia and still use spell check. Anyone can add footnotes to make something look more serious (students do it all the time!) The same holds for using scientific language.

And a lot of things students are told to look for have no basis in reality at all. In our research we have found students believe