Designing Homework That Enhances Learning

What kind of homework assignments promote learning? We don’t need research to confirm that doing homework benefits most (maybe it’s all) college students. But there are some vexing issues. If the homework is graded and if those grades count, students will do the homework. But then all that homework must be graded. That can involve a huge time investment for the teacher. So, faculty respond by designing homework assignments that can be graded quickly or aren’t graded at all, with students getting credit for completing them, provided the work shows they’ve made a reasonable effort. Both of those options tend to compromise the amount of learning that results from doing the homework assignment.

A faculty research team tested an interesting homework design feature in multiple sections of an undergraduate educational psychology course. Homework was assigned in every class session, and it consisted of 10 to 12 short-answer essay questions. Even though most of the questions could be answered in two to five sentences, that’s enough homework grading to bury most teachers. In one of the experimental conditions, students simply got credit for the percentage of questions answered in each homework assignment—the completion option. In the other, answers to 10 percent of the questions (randomly selected) were graded according to a quality criteria explained in the article. In this option students knew some of their answers would be graded, they just didn’t know which answers. The second approach reduced the instructor’s grading time by 90 percent. Credit for both assignment options was the same.

Researchers were interested in two questions. “The first goal of the study was to determine if a randomized credit contingency [the second option] would produce higher-quality answers than assessment of all homework items for completion.” (p. 65) “The second goal was to see if improvement in the quality of students’ homework would indirectly improve their exam performance.” (p. 65) And the results confirmed both these hoped-for conclusions. “Setting randomized reward contingencies specifically for accuracy of homework produced both significantly higher accuracy and length of homework answers than a reward contingency based on completion of the homework.” (p. 73) The randomized answer option was also associated “with modest but significant gains in adjusted exam scores.” (p. 73)

In a discussion of how these findings relate to other research, the study authors note that over the past decade, research has identified a number of factors that either contribute to or actually predict exam scores. They list (and reference) homework completion, critical-thinking ability, participation in class discussion, generic vocabulary, and student efficacy. “None of these studies considered separately constitute a dramatic contribution to exam scores, but taken together they provide a relatively extensive picture of what accounts for exam performance.” (p. 74)

Completing a homework assignment like this one not only promotes the learning of course content, but it also brings students to class prepared to talk about the content. They can answer questions and add insights, and that makes for richer class discussions.

Design details, even small ones, do make a difference. In this case they encouraged students to prepare longer and more accurate answers to questions about course material, and those answers were associated with better exam scores. And by reducing the number of answers that were graded, this design detail makes giving regular and substantial homework assignments a viable option for faculty members. Would students object to having to prepare answers for which they got no credit? In this case, exam questions were conceptually linked to the homework assignment questions. An example in the article illustrates this connection. If students are shown the relationship between homework questions and those on the exam, that may dampen their objections.

Reference: Galyon, C. E., Voils, K. L., Blondin, D. A., and Williams, R. L. (2015). The effects of randomized homework contingencies on college students’ daily homework and unit exam performance. Innovative Higher Education, 40 (1), 63-77.

 

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Related Articles

Love ’em or hate ’em, student evaluations of teaching (SETs) are here to stay. Parts <a href="https://www.teachingprofessor.com/free-article/its-time-to-discuss-student-evaluations-bias-with-our-students-seriously/" target="_blank"...

Since January, I have led multiple faculty development sessions on generative AI for faculty at my university. Attitudes...
Does your class end with a bang or a whimper? Many of us spend a lot of time...

Faculty have recently been bombarded with a dizzying array of apps, platforms, and other widgets...

The rapid rise of livestream content development and consumption has been nothing short of remarkable. According to Ceci...

Feedback on performance has proven to be one of the most important influences on learning, but students consistently...

wpChatIcon
What kind of homework assignments promote learning? We don't need research to confirm that doing homework benefits most (maybe it's all) college students. But there are some vexing issues. If the homework is graded and if those grades count, students will do the homework. But then all that homework must be graded. That can involve a huge time investment for the teacher. So, faculty respond by designing homework assignments that can be graded quickly or aren't graded at all, with students getting credit for completing them, provided the work shows they've made a reasonable effort. Both of those options tend to compromise the amount of learning that results from doing the homework assignment. A faculty research team tested an interesting homework design feature in multiple sections of an undergraduate educational psychology course. Homework was assigned in every class session, and it consisted of 10 to 12 short-answer essay questions. Even though most of the questions could be answered in two to five sentences, that's enough homework grading to bury most teachers. In one of the experimental conditions, students simply got credit for the percentage of questions answered in each homework assignment—the completion option. In the other, answers to 10 percent of the questions (randomly selected) were graded according to a quality criteria explained in the article. In this option students knew some of their answers would be graded, they just didn't know which answers. The second approach reduced the instructor's grading time by 90 percent. Credit for both assignment options was the same. Researchers were interested in two questions. “The first goal of the study was to determine if a randomized credit contingency [the second option] would produce higher-quality answers than assessment of all homework items for completion.” (p. 65) “The second goal was to see if improvement in the quality of students' homework would indirectly improve their exam performance.” (p. 65) And the results confirmed both these hoped-for conclusions. “Setting randomized reward contingencies specifically for accuracy of homework produced both significantly higher accuracy and length of homework answers than a reward contingency based on completion of the homework.” (p. 73) The randomized answer option was also associated “with modest but significant gains in adjusted exam scores.” (p. 73) In a discussion of how these findings relate to other research, the study authors note that over the past decade, research has identified a number of factors that either contribute to or actually predict exam scores. They list (and reference) homework completion, critical-thinking ability, participation in class discussion, generic vocabulary, and student efficacy. “None of these studies considered separately constitute a dramatic contribution to exam scores, but taken together they provide a relatively extensive picture of what accounts for exam performance.” (p. 74) Completing a homework assignment like this one not only promotes the learning of course content, but it also brings students to class prepared to talk about the content. They can answer questions and add insights, and that makes for richer class discussions. Design details, even small ones, do make a difference. In this case they encouraged students to prepare longer and more accurate answers to questions about course material, and those answers were associated with better exam scores. And by reducing the number of answers that were graded, this design detail makes giving regular and substantial homework assignments a viable option for faculty members. Would students object to having to prepare answers for which they got no credit? In this case, exam questions were conceptually linked to the homework assignment questions. An example in the article illustrates this connection. If students are shown the relationship between homework questions and those on the exam, that may dampen their objections. Reference: Galyon, C. E., Voils, K. L., Blondin, D. A., and Williams, R. L. (2015). The effects of randomized homework contingencies on college students' daily homework and unit exam performance. Innovative Higher Education, 40 (1), 63-77.