I didn't say it. It comes from a recent Alertbox post from Jakob Nielsen. He is a big name in the web usability world, and not an educator. But I thought his ideas were interesting and I wonder if we might see similarities with our own students - particularly those learning either fully or partially online for courses.
One of his ideas is that people remember much more after reading if they are asked to retrieve information about the text from memory using a test. Nielsen cautions that he has long thought that the Web may be "unsuited for real learning" mostly because of the superficial way that users "surf" (the "scan" of the digital age) information instead of close reading.
His own usability studies on teens leads him to conclude that "When using websites, teenagers have a lower success rate than adults and they're also easily bored. To work for teens, websites must be simple -- but not childish -- and supply plenty of interactive features."
College students don't rate much better because they are "multitaskers who move through websites rapidly, often missing the item they come to find. They're enraptured by social media but reserve it for private conversations and thus visit company sites from search engines."
Again, Nielsen is looking most of the time at non-educational (commercial) uses of the Web.
Most of his post is about "writing-for-the-Web" guidelines, and it would seem that the thrust is at "making it simple." (Examples: "Craft the first 2 words of headlines and other microcontent for the scanning eye; Use numerals instead of words; Use bulleted lists") That's probably not an approach that appeals to educators.
Nielsen recognizes that, and sees the overlap here of educational sites where users need to learn something beyond the highlights, and commercial sites that need to "educate" users. (For example, a pharmaceutical site for patients about a drug.)
He cites a recent research study by Karpicke and Blunt from Purdue University published in the January Science that measured the amount of information people could remember a week after reading a scientific text. The students who took an "elaborate" test after reading the text remembered 145% more content after a week compared to the untested students who simply read the text.
I found it interesting that people who simply read the text 4 times also remembered more (64%) a week later.
Nielsen is concerned with the implications for web designers and I think this is probably an area that must also concern instructional designers working on web content for courses.
Is quizzing and testing (high and low stakes) the way to get better retention?
Unfortunately, for commercial web designers, they can't usually "require" users to take tests. But teachers often can and do require objective and subjective tests.
Perhaps, Nielsen's final thought is more important to our course design thinking - "get users to exercise their memory after reading your content, and then offer them a chance to revisit the material after they've seen how little they remember."