Retrieval Practice in the Classroom: Is Asking Questions Enough?
By Cindy Nebel
In our workshops, we often recommend that instructors start by making small changes in their teaching, instead of trying to do a massive overhaul to a curriculum or design from scratch. One possible way that an educator may choose to incorporate retrieval practice into their classroom is to simply stop and ask questions throughout a traditional lecture. By asking students questions, they will all be in engaging in retrieval practice without the instructor taking too much time out of their class to develop or grade quizzes. But is that enough? In today’s post, I review a study that asked just that question and has implications for student group work as well.
In this study (1), Magdalena Abel and Henry Roediger had students study Swahili vocabulary words in pairs. First, both students sat at a computer while the Swahili word was presented with its English equivalent (study phase). Some of the words were shown again, giving the students an opportunity to restudy, but for other words, the students would take turns trying to guess the English word out loud when presented with the Swahili word (retrieval practice). While one student was engaging in retrieval practice, the other student was grading them in one of two ways: either they were graded for accuracy or for fluidity. In the accuracy condition, the partner was required to recall the correct answer themselves in order to grade their partner’s response. In this way, they were still engaging in retrieval practice, but what we call covert retrieval practice. In the fluidity condition, the partner was told to grade their partner on how smooth and fluid the response was. Participants then took a final test over all the vocabulary words after either 3 minutes or 2 days.
Surprisingly, covert retrieval practice simply didn’t work. On the final test 2 days later, students actually performed worse on the items that were studied with covert retrieval (by grading their partner) than on the restudied items.
One possibility is that the students who were grading their partners didn’t actually engage in effortful retrieval. Instead, they may have waited for their partner’s response and then judged whether or not it seemed accurate. So, in a second experiment, Abel and Roediger specifically told students to try to recall the correct answer, but there still wasn’t much difference between covert retrieval and restudying.
The authors speculated that the partners still might not be putting forth much effort to try to retrieve the answers themselves, so in their third experiment, they asked the partners to grade their own covert retrieval. In this way, students needed to recall to grade themselves, further pushing them to engage in covert retrieval practice. In this final experiment where students were not just grading a partner, but also monitoring their own retrieval, they finally found an effect of covert retrieval practice producing better retention than restudying.
So, what does this mean for educators?
It is not enough to ask questions during lecture because students are unlikely to engage in covert retrieval unless they are pushed to do so. Instead, try having students write down an answer and then possibly share that answer with a partner or the class, but they need to engage in their own retrieval first. Simply asking students to engage in retrieval is not enough as many will likely wait for the answer and then engage in a recognition process (“Oh yeah, I knew that”) instead of recalling the response for themselves.
The other important takeaway more directly impacts student group work. When students are engaging in retrieval practice in groups, the person who is in charge of asking the questions would benefit from also writing down a response or they will see little benefit to this group work. Alternatively, it is important that students take turns frequently during group work, making sure that everyone gets asked the same questions so that all students benefit from the retrieval on each question.
(1) Abel, M., & Roediger III, H. L. (2018). The testing effect in a social setting: Does retrieval practice benefit a listener?. Journal of experimental psychology: Applied, 24(3), 347.