Homework Problems (or Problems with Homework)
A topic that comes up again and again among my friends and colleagues is the homework question. How much to assign? How frequently to assign? What kind of problems should we assign? Does online homework work?
Kukliansky et. al. (2014) report that 20% of teachers don’t find homework valuable. I’m assuming that’s K-12 teachers, but that’s a lot of people questioning the value of homework!
In the literature review, Kukliansky outlines three advantages to homework:
Cognitive – it helps with student learning and achievement
Pedagogical – it helps students to prepare for class, and participate more fully during class time
Affective – it helps develop independence and responsibility
The disadvantages are:
Affective – an increase in workload can cause students to disengage
Social – too much homework interferes with out of school activities
Pedagogical – not enough guidance from teachers can cause confusion
In my experience I have found that the traditional weekly problem sets weren’t working for the introductory courses. Students would leave the assignment until the day before it was due, and then either camp out in my office all day or else stay up all night working on it. Then they were tired and useless for class the next morning (one of my intro classes is at 8:00 am).
Additionally, it didn’t allow me to give feedback in a timely manner. By the time I had corrected the homework and passed it back, it easily could have been two weeks since we covered that material in class. This meant a lot of backtracking to correct issues that the students were having with the concepts or problem solving strategies.
As a consequence of this, I’ve moved to assigning a small amount of homework each night.
Typically one or two problems and one or two conceptual questions. I tell the students to spend no more than an hour on the assignment, which helps them to manage their time more effectively. If they find they are spending more than an hour on the assignment, they should see me or a tutor, and not waste any more time beating their heads against the desk. It also shortens the feedback loop. Because the assignments are short, I can usually grade them between each class meeting, which gives me a good idea of where the students are having trouble and allows me to correct course immediately. At first, I was worried the students would push back because of the increased frequency of assignments, but it has worked out remarkably well. The feedback from students has been overwhelmingly positive, especially from those who previously did the long weekly assignments. The students by and large recognize that they should be spreading their studying out over the week, but have difficulty doing it on their own.
So that has been my experience, but I was curious what the research had to say about homework. Turns out that there is not a huge amount of research on the effectiveness of homework at the college level. Most of what I found at the college level had to do with either online homework assignments or the role of homework in flipped classrooms. (I decided not to get into the value of flipped classrooms here because that’s a whole post on its own. We’ll focus for now on more or less traditional homework assignments.) It is difficult to generalize the findings because the studies are always specific to a particular class in terms of discipline, size, pedagogy used, etc. In other words, it is difficult to do a controlled study. In short, the results are mixed. In their review of 69 studies between homework and achievement, Cooper et al (2006) found that 50 studies had a positive correlation and 19 were negative. Here’s a few of the highlights that I found:
Planchard et al (2015) found that the top motivating factors for students to complete homework in a Genetics course was reinforcement (28.75%) followed by credit (18.75%). This indicates that at least some students are completing the homework for the reasons we want – to learn the material, but many are just doing it because they have to. Only 6.25% reported doing it because they were interested. On the other hand, this study found time management was the biggest barrier to students not completing homework (27.85%) following by not understanding the questions and difficulty. This means that we as teachers need to be clear about our expectations for time spent on out of class assignments, and to make sure that the students have the tools they need to be able to complete the assignments (both in understanding the assignment and the conceptual content).
Ryan and Hemmes (2005) compared for-credit and no-credit homework completion in a Psychology class, and found that students in the no-credit group did not continue completing assignments throughout the semester. This is a kind of a “duh” result, but it’s always good to see a concrete piece of evidence for what we anecdotally know to be true – students won’t complete assignments if they don’t count for something. And I did make this mistake as a beginning teacher. Because I was worried that grading would take over my life, I tried not requiring homework for a semester. I would assign problems to do because they would help them learn the material, but not grade them. No surprise, they didn’t do them, and as a consequence didn’t learn the material. So then I started giving “take-home, open-book quizzes” and all of a sudden – boom – they did the assignments. It’s all about framing.
A review article by Cooper et al (2006) finds that there is “consistent evidence for a positive influence of homework on achievement” particularly for grades 7-12. Their review did not explicitly look at homework at the college level. (Interestingly, homework has no correlation to achievement in the elementary grades.) However, he goes on to state “when compared with in-school supervised study, random-assignment designs revealed no difference between the homework and in-school study students.” This indicates that the value is in practice, but that can be done in the classroom as well as at home. I think this another indicator that we need a post on flipped classrooms, which have proliferated since Cooper’s article was written.
Planchard et al (2015) also found a strong correlation between homework completion and exam scores in a college Genetic class, confirming the results from the K-12 research. They did not comment on whether or not completing the homework correctly was correlated to higher exam scores. Is trying and failing just as effecting as getting it right the first time?
Lazarova (2015) found that online HW helped her non-physics majors learn problem solving skills. What’s interesting about this study is that they are using online homework at a small, liberal arts college, and finding it to be more effective than traditional paper-based assignments. (Most other studies on online homework are at big universities.) They had noticed that homework grades were not strongly correlated with exam scores – students could do well on homework and still not succeed on exams. However, when they switched to online homework (WebAssign), then homework grades became a good indicator for exam scores. The type of problems assigned in both cases were similar, but the online homework tool allowed for multiple attempts and hints to help along the way. A study of chemistry courses found similar results – the adaptive online homework that gives immediate feedback and assistance was more effective than the traditional responsive homework (Eichler & Peeples, 2013).
Kontur et al (2015) found that homework completion was correlated with increased exam scores only for high aptitude students. They offer suggestions that this might be due to cognitive overload, ineffective homework strategies, or mismatched assessments (between homework and exams).
I must say that I have been skeptical of online homework programs, but research like this makes me think that there is value in this new-fangled technology. One question I have about both of these studies is regarding time on task. One effect of the online homework seems to be a motivation to keep trying the problem until they get it right, which means they may be spending more time on the assignments, or at least learning to be more persistent.
Another concern with online homework is that the teachers might not look closely at the results. One reason to assign homework is to assess our teaching, and find out if there are topics we need to revisit and improve instruction. But that only happens if we correct the homework ourselves, and not outsource it to a computer or a TA.
Kukliansky (2015) found that a shocking 40% of teachers made no reference to the HW assigned during class. This is a distressing result. If we assign students to do a reading or assignment, we need to make sure that it is integrated into the lesson. This is particularly true of preparatory assignments. If they don’t need to have done the reading or assignment before coming to class, then why should they do it? This is a good way to disengage students. Follow up assignments (like problem sets) also need to be addressed in class, especially if there are concepts that the students didn’t get right.
As I wrap this up, I’m thinking about how closely homework is intertwined with the pedagogy we use in the classroom. Which is why so much of the literature on homework these days is related to “flipped classrooms.” I’ll have to make that the subject of a future post.
Cooper, H., Robinson, J., & Patall, E. (2006). Does homework improve academic achievement? A synthesis of research, 1987–2003. Review of Educational Research, 76, 1–62.
Eichler, J. & Peeples, J. (2013). Online Homework Put to the Test: A Report on the Impact of Two Online Learning Systems on Student Performance in General Chemistry. Journal of Chemical Education, 90, 1137-1143.
Lazarova, K. (2015). The Role of Online Homework in Low-Enrollment College Introductory Physics Courses. Journal Of College Science Teaching, 44(3), 17-21.
Kontur, F. LaHarpe, K., Terry, N. (2015). Benefits of completing homework for students with different aptitudes in an introductory electricity and magnetism course. Physical Review Special Topics – Physics Education Research, 11, 010105.
Kukliansky, i. Shosberger, I. & Edhack, H. (2016). Science teachers voice on homework: beliefs, attitudes, and behaviors. International Journal of Science and Mathematics Education, 14 (Sup 1): S229-S250.
Planchard, M., Daniel, K. L., Maroo, J., Mishra, C., & McLean, T. (2015). Homework, Motivation, and Academic Achievement in a College Genetics Course. Bioscene: Journal Of College Biology Teaching, 41(2), 11-18.
RYAN, C.S., AND N.S. HEMMES. (2005). Effects of the contingency for homework submission on homework submission and quiz performance in a college course. Journal of applied behavior analysis 38(1): 79–88.