GUEST POST: How to Shift a School Towards Better Homework
By Ian Kelleher
Dr. Ian Kelleher is the co-author of Neuroteach: Brain Science and the Future of Education. He teaches chemistry, physics and robotics, and coaches boy’s JV soccer, at St. Andrew’s Episcopal School in Potomac, MD, Ian is also Head of Research for the school’s research group, the Center for Transformative Teaching and Learning (the CTTL). Ian’s work for the CTTL focuses on helping teachers translate Mind, Brain and Education research into classroom practices, and measuring the impact. He is also responsible for the CTTL's professional development programming. Ian has presented to schools and conferences throughout the world, including the Learning & the Brain Conference, SXSWedu, the International Mind, Brain, and Education Science Conference, the Daily Telegraph Festival of Education, ResearchED, and the National Association of Independent Schools Annual Conference. Ian is part of the CTTL team that was awarded the International Mind, Brain and Education Society's Mission Award in 2016.
Talk to any high school student about homework, and then watch their face.
Do we do homework well? How can research help us do it better?
Parents and teachers mistakenly equate hours of homework with rigor, the sheer quantity of often tedious hours late into the night with academic excellence. What if a school commits itself to using research to craft a different path? Can we use research to create greater, deeper learning in fewer hours? And, in doing so, can we help create happier, less sleep deprived, less anxious students?
Our homework began with a conversation with students on the way home from a field trip. St. Andrew's Episcopal School, Potomac, Maryland has, for ten years, been using research to inform practice, but students were frustrated that we had not tackled homework. We realized that our response of, “But that is soooo hard!” was not good enough.
We had to act. But how do you shift this most stubborn of bars? What if students became masters of the research, and then presented to the faculty? There is a power in connecting research evidence to professional practice and pride in being masters of our craft, and even greater power when we then connect it to the faces and stories of real students who will be affected by what you do. We aimed to leverage this.
We firmly believe that students should be involved in a school’s research process - their voice is critical. At St. Andrew’s, we had already appointed Student Research Fellows, who work alongside Teacher Research Fellows and the Head of Research at the school’s research group, the Center for Transformative Teaching and Learning (CTTL). A previous project by our student fellows on the relationship between happiness, motivation, and achievement led to this story in Time Magazine. Our student fellows were the perfect ones to take on this project.
Our student fellows wanted to rush into action, do an intervention, and see how it worked - they wanted change! But we convinced them to start with an ethnographic case study, a type of research that can work well in schools and is an important step to take before running a full pre/ post-test intervention study. It worked like this:
- Create a survey instrument to collect quantitative and qualitative data. Make it short enough and interesting so that students would be interested in completing it carefully. Significant student input here makes a dramatic improvement in the quality of the survey tool. You can view the survey our students created here.
- Conduct the survey and analyze the results.
- Compare the results to published research (see this review (1), this white paper, and this Education Endowment Foundation resource for the sources we used with students, chosen because they are a good balance of research evidence and accessibility).
- From this, craft recommendations
- Test the recommendations in focus groups, and iterate the process
- Students present the recommendations to the faculty
How do you begin shifting the way teachers set homework? There is immense power in seeing students speaking confidently and knowledgeably in front of the entire faculty about their data on homework, published research, and how this all fits together with their own and their peers’ lived experience.
There is even greater power in opening the floor up for questions and having the students field them thoughtfully, confidently, and effortlessly, citing research in their answers. Our student fellows now know more about the field of homework research than do our faculty. This point is not lost on our faculty. The sacred cows of past practices suddenly seem less justifiable. It is time to act. But what should teachers do? We have four simple suggestions, and one harder one.
1. Continue setting homework
In free-response answers to the question, “In an ideal world, how many hours would be most beneficial?” only 9% of our students wrote in answers between zero and one hour for the ideal total amount of homework per night. In addition, 51% suggested 1 to 2 hours, 31% 2-3 hours, and 9% more than this. Our students seem to believe in the value of homework. This agrees with the research that homework can aid learning (see this Education Endowment Foundation resource). Our hypothesis is that students have a positive attitude to homework experiences that help them learn…
2. But, work harder to make it meaningful
…as long as these experiences appear meaningful. We found that teachers overestimate how good a job they are doing of expressing the purpose in each assignment they set. Why does this happen? Here are some ideas:
- the curse of the teachers’ (deep) knowledge of the subject. Teachers often underestimate the degree of scaffolding needed to help students transfer knowledge and skills to new contexts, and overestimate students’ ability to do this automatically themselves (see the Top 20 Principles from Psychology for preK-12 Teaching and Learning). It is our belief that teachers often overestimate how clear the purpose of an assignment is – but the good news is that we believe a small amount of deliberate effort by teachers here will go a long way.
- homework is often talked about in a rush at the end of class, if at all;
- we just don’t realize how massively important purpose is to students – it is one of the ‘big three’ mindsets in education, but not one that most teachers or school leaders think of.
There is a large amount of research on the role of purpose in learning (see this brief), and these simple-to-institute and potentially transformative strategies should not surprise us:
Before setting homework:
Make sure your assignments are meaningful - which could be to the outside world, for an upcoming assessment, or for something that they will soon do in class; and then explain this meaning to students.
After setting homework:
Make sure that homework ties back into the story of the class, or is checked - make sure it does not disappear into a void. As one of our student fellows said, “if it is hard to tie the homework into the story of the class, perhaps it wasn’t good homework.”
3. Gives students appropriate choices
Where possible, offer choice in topic to increase student motivation. Making room for students’ creativity, and fostering each child’s creative development, can increase engagement (see the Top 20 Principles from Psychology for preK-12 Teaching and Learning).
More challenging is working with students to develop their ability to self-monitor and choose when they stop in a list of review questions. They will need coaching on how to do this because they will tend to stop too soon – persisting beyond the ‘I get it now’ point in retrieval practice is important. However, boredom is actually linked to stress (2), and I believe that teachers underestimate the frustration and resentment this breeds (see this Education Week piece). It is very much worth teaching kids how to stop at the right time - you are helping to build precious metacognitive and self-regulation skills (see the Top 20 Principles from Psychology for preK-12 Teaching and Learning).
Leading homework researcher, Professor Harris Cooper of Duke University, coined the “10-minute rule” as a beautifully simple and workable summary of his complex research (3). It suggests optimum amounts of 60 minutes for 6th graders, 90 minutes for 9th graders, and two hours for seniors, and so on for total amounts of homework per night. Beyond this, Cooper says, homework has a negative impact on learning. The ‘ideal’ amount of homework our students suggested in our survey match Cooper’s findings well. As do the negative consequences they reported: sleep deprivation, anxiety, loss of time with family, and hampering their ability to do things beyond school. Homework can have a significant negative effect on student wellbeing (4); see this Stanford University press release.
For effective homework, quality is much more important than quantity. To hammer this point home, I urge all educators to read this short piece from Stanford professor Denise Pope’s Challenge Success group. We must resist our self-important urges as teachers, which sometimes push us to assign over-ambitious amounts of homework.
5. As a school, have a strategic grading policy
We believe this statement from one of our juniors speaks for many students, “I will do the homework that gets me the most points, even if I totally understand it, and not the homework that doesn’t get any or as many points, even though I need to study it more.” This is the homework game, where the focus is on survival, not learning. Minimizing the homework game takes grade-level wide cooperation within a school.
The homework game
Our students play out the homework game each night. For example, take a look at the beautifully completed active reading assignment in the image below.
It was, in her words, “faked” by one of our students – done in a couple of minutes so she could clear it from her to-do pile and move on to other tasks. It is a skill she tells us she has mastered over many years practice. As a junior, she feels she is now able to give the impressions of having done conscientious, thoughtful homework in a matter of minutes without any deep engagement in the task. She tells us she has figured out exactly what she needs to do to get points and move on to the rest of her homework pile. It is unlikely that this student learned anything from the activity, as she was simply in survival mode.
How many times is this game played out each night across the country? Our student was initially reluctant to share her story, feeling a bit like a whistleblower, in danger of running ‘the game’ for all students everywhere. She agreed to share her story – and has even done so in public – because she believes that doing so can help teachers set better homework.
We need to stop perpetuating a system which leaves students anxious, miserable, and sleep deprived, often for marginal gains in learning. The pain versus gain is not worth it. Parents should not have to choose between academic rigor and wellbeing for their child; they should, and can, have both.
Thanks to our student research fellows, we were able to make this point to our faculty in a way that created emotional resonance. The students’ recommendations seemed fair and well argued - and failure to act on these simple recommendations is now tied to real faces, people they care about and want to do the best for. We all left that room imagining a future where learning for all could be improved while, with the same actions, anxiety could be taken down a notch. That is the kind of change that all schools should be aiming for.
(1) Cooper, H. (1989). Synthesis of research on homework. Educational Leadership, 47, 85-91.
(2) Merrifield, C., & Danckert, J. (2014). Characterizing the psychophysiological signature of boredom. Experimental Brain Research, 232, 481-491.
(3) Cooper, H., Robinson, J. C., & Patall, E. A. (2006). Does homework improve academic achievement? A synthesis of research, 1987–2003. Review of Educational Research, 76, 1-62.
(4) Galloway, M., Conner, J., & Pope, D. (2013). Nonacademic effects of homework in privileged, high-performing high schools. The Journal of Experimental Education, 81, 490-510.