Pre-testing is an educational strategy that involves giving students ungraded tests before they’ve had a chance to learn material. Though it can sometimes feel counterintuitive, it has a continually growing body of research supporting its use.
Steven C. Pan, director of the Learning Sciences Laboratory at the National University of Singapore, is a leading pre-testing researcher, and has published a series of recent studies on pre-testing that highlight its benefits as well as some of its limitations.
Pre-Testing: The Benefits of Guessing
Earlier this year, Pan co-authored a study that examined pretesting in the context of language learning. Pan notes that popular language learning apps often ask users to guess what they think a word means without necessarily having any idea of the word’s meaning.
“Language learning software developers, they often will say that this is something that they feel is research-related in terms of it's possibly an effective way to learn,” Pan says. “We put that to the test not too long ago, and we found that using this form of guessing, practice testing with correct answer feedback would actually enhance learners' ability to remember the meanings of words, to recognize words, and things like that.”
The research was another indication that pre-testing can be an effective tool for learning.
Pre-Testing and Memory
Pan’s lab also recently published a paper that looked at pre-testing and memory/retention. For the research, they split participants into multiple groups. Some learned with pre-testing, some without. Then, Pan and his fellow researchers measured how much each participant remembered over time.
“For some people, the memory test happened five minutes later, some it happened a day later, some it happened a week later, two weeks later, and in one of the experiments, it happened a month later,” Pan says. “So we got to see the retention or forgetting curve over time.”
They found that participants who were introduced to pre-testing learned more material initially but forgot it just as quickly as those who did not experience pre-testing.
“Right away, within five minutes, performance is much, much higher. So there's a boost in recall, and then the forgetting rate is the same as the other conditions,” Pan says.
However, because they learned so much more initially, a month later, for instance, those who learned with pre-testing still were ahead of those who had not learned with pre-testing.
This study suggests both the potential and limitations of pre-testing, Pan says, suggesting the method can help you learn something really quickly. “But if you want to remember it really well a month from now, you might have to do some additional things after that,” he adds.
Using Pre-Testing In The Classroom
Though there is still more to learn about pre-testing, Pan says the evidence is strong enough that teachers can go ahead and utilize it in their classes and don’t need to worry about potential downsides.
“Sometimes, teachers will say, ‘Well, if we haven't taught them anything, and the idea is to make them guess, aren't they gonna produce a bunch of wrong answers? And isn't that problematic that they might start to think the wrong thing?’” Pan says. “In our studies, and studies in the literature more broadly, as long as the students have a chance to learn the right answers, and actually get a lesson from the teacher, or read the textbook, or watch a video that provides them with the correct information, and they're paying attention, there doesn't seem to be any problem.”
Pan says that as long as you keep the tests low-stakes and make sure to include the correct solutions, pre-testing is a great strategy for teachers, and one that doesn’t cost anything or require a lot of extra work.
“It can be quite easy in practice, I think, to just pose a couple of questions about key topics or ideas that the teacher might think is important, and they want their students to really master, and just get them thinking about it before the lesson or at the start of the class.” He adds that teachers might also want to consider pairing it with post-testing and retrieval practice methods. “That might be the nice way to sort of bookend learning from both ends.”
