The Science of Better Homework Assignments

Though I have long been fascinated (cautiously, though, I hope) by the potential for applying the mind sciences to the study of history, I have not given as much attention to new educational psychology research.  This week, however, I learned of a new Duke and Rice study that suggests some easy but effective ways of designing homework that helps students perform better on tests.  Of course, test performance is meant to be only a means to a learning outcome, not an end in itself.  Also, the study in question looks at students of engineering rather than of history.  Still, the principles are interesting.

In the study, the usual homework in an engineering course was alternated each week with assignments that stressed the principles of repeated retrieval, spacing, and feedback. Instead of simply moving onto a new type of problem each week, students were given follow-up problems through two additional assignments that asked them to retrieve lessons they’d already learned (repeated retrieval).  Instead of receiving problem sets in one assignment, the problems were given out over three weeks (spacing). And the instructor responded immediately to student work, rather than handing back critiques a week later (feedback).  The study split the student body into two groups, alternating which group was receiving the modified assignments just discussed.  In the final exam at the end of the semester, each group received a 7 percent higher grade on material explored in the modified assignments than did the group that had completed more conventional homework assignments on that material.

I find these results impressive, and though certain aspects of the method (feedback in particular) might be difficult to replicate in the history classroom, others may be less so: one might for instance encourage repeated retrieval and spacing simply by asking some of the Week 7 discussion questions during Weeks 8 and 9.  This is, however, something we rarely do.  In theory, we instructors try to use the concept of the “theme” in order to encourage students to intellectually revisit and build upon earlier discussions throughout our courses.  In fact, I’m not sure this always works as well as it could; for one thing, undergraduates are much less accustomed to thinking about themes than professional historians.  In many history courses, there is little opportunity for repeated retrieval or spacing, as weekly discussions tend to continue the historical narrative without genuinely returning to earlier “problems.” Thus, I’d like to present the question: how do you encourage students to re-visit earlier material without confusing them or impeding the forward progress of the course?