By Megan Sumeracki

Like with many effective learning strategies, what students think is helping them learn is not what actually helps them learn. In two experiments presented by Hillary Mullet and colleagues (2014, 1), University engineering students received relatively immediate feedback or delayed feedback on homework assignments. I talked about this paper in a bite-size research episode on our podcast; so, if you prefer to listen you can head to our podcast episode list and check out Episode 58. But, for those who prefer to read or want a different modality (with spaced repetition, albeit a pretty long space…) read on.

If you just want a quick summary, scroll to the bottom and read the bulleted list of results and the closing paragraph.

If you’re reading this blog, then you probably care about learning. A lot. If you’re a teacher, you want to foster as much learning in your classroom as possible. As a part of this, you probably try to give your students as much feedback as possible. You also probably have the sense that providing feedback as quickly as possible after an assignment is best. However, immediate feedback can be really difficult, if not impossible, to do. If you’re a student, you likely prefer immediate feedback to waiting for it. As a student, you may or may not have control over when feedback is delivered.

Historically, the idea that feedback should be immediate to best improve learning seems to come from behaviorism. Mullet and her coauthors provide a bit of history on this and quote Skinner. He suggested a delay of 24 hours to receive feedback was too long. (24 hours?! At least in my world, that’s fast.) Logically if we think about feedback as correcting errors, then it makes sense that we would want pretty immediate feedback. But if we think about feedback as another presentation of the information, then a space ought to improve learning.

Mullet and colleagues tested whether feedback was best immediately (or, at least more quickly) or after a delay. What I love about this paper is that they ran their experiments in real Engineering classrooms, which means the “immediate” feedback condition is at least sometimes practical. By immediate, they mean that the feedback was available immediately after the assignment deadline. So, not necessarily immediately after the students completed the assignment, but more quickly than in the delayed condition.

Experiment 1

The experiment was conducted in real engineering classrooms. The students attended weekly lectures, had course material, like videos, to digest at home, and had weekly homework assignments through an online e-learning system to complete.

The researchers manipulated when feedback for the homework assignments was available for students through random assignment. Some students were able to view the feedback immediately after the due date for the homework assignment, while others were able to review the feedback one week after the assignment was due.

The students were required to review the feedback to earn course credit, so we know they are at least opening the feedback! The researchers reported when and for how long the students in each condition viewed the feedback. Of course, making the feedback available immediately after the due date does not mean that students reviewed the feedback immediately. On average, those in the “immediate” feedback condition viewed the feedback 4 days after they completed the homework, whereas those in the delayed feedback condition viewed the feedback 12 days after they completed the homework. Interestingly, students in the “immediate” feedback condition viewed the feedback longer than those in the delayed feedback condition. So if anything, this means the experiment is slightly biased towards the “immediate” feedback condition.

The researchers then looked at performance on the class exams. Questions on the exams required the students to transfer what they had learned to novel problems. They also asked the students how well they thought the timing of the feedback helped them.

 Experiment 1 Results

Students in the delayed feedback condition performed better on the exams than students in the “immediate” feedback condition. The difference was about 8%, or the difference between earning a B and an A. However, students’ perceptions did not match the exam results. The students that received the delayed feedback really didn’t like it. They either thought it did not help them learn or even thought it hurt their learning. Students who got the more immediate feedback liked it and thought it helped them.

Experiment 2

Mullet and colleagues conducted a second experiment to replicate the results from the first and made a couple of small changes. Experiment 2 was largely the same as Experiment 1, but there were two important changes. The first was that they manipulated the timing of feedback within students. On some homework assignments, students received “immediate” feedback (i.e., it was available immediately after the due date), and on others, they received delayed feedback.  Manipulating the timing of feedback in this way allowed the researchers to see how students compare the two types of feedback after they have experienced both. Perhaps after experiencing both types of feedback, they will realize that the delayed feedback was actually better for their learning and subsequent exam performance.

The second change was that they manipulated the feedback requirement. Some students were required to review the feedback for course credit, like in Experiment 1. However, for some students, viewing the feedback was optional.

Experiment 2 Results

The results from Experiment 2 were very similar to those from Experiment 1, even with some methodological changes! Here is a summary of their results:

• Delayed feedback led to better exam performance than “immediate” feedback. This was true even when the feedback was optional!

• Overall, requiring feedback led to better exam performance than making the feedback optional.

• Students still report that the immediate feedback feels like it is helping them whereas the delayed feedback feels like it is hurting them, even when they experience both in the course. In fact, 79% said they benefited a lot more from the immediate feedback.

This paper joins a number of other papers showing delayed feedback is beneficial. The finding is good news, practically, for teachers. Relax! There’s no need to rush to get feedback to students, and they will benefit from the extra time it often takes. However, you may have to convince your students that this is benefitting them. I will say, as a teacher, learning scientist, and someone who is always trying to learn myself, it is sort of annoying that our intuitions about what helps us learn are so often wrong!

 References

(1) Mullet, H. G., Butler, A. C., Berdin, B., von Borries, R., & Marsh, E. J. (2014). Delaying feedback promotes transfer of knowledge despite student preferences to receive feedback immediately. Journal of Applied Research in Memory and Cognition, 3, 222-229. http://doi.org/10.1016.j.jarmac.2014.05.001