Reflections on The Homework Myth with Alfie Kohn

Education is like a tapestry. Wiggle one thread and many others will move as well.
Alfie Kohn

Following on from his lecture on The Case Against Competition, which I reflected upon here, and a break for dinner, Alfie Kohn delivered a lecture on The Homework Myth, As much as I struggled to reconcile some of what Alfie said during his Competition lecture, much of what was said during this lecture mirrored a lot of my own thoughts on homework.

The lecture was introduced using the above quote and then Alfie spoke about what would happen if we abolished grades. Students, we were told, would question the point of why they attend school and completed the tasks they are assigned. This means that we, the teachers, need to reconsider what and how we teach; to reexamine the curriculum and the pedagogy. This is an interesting challenge and one that seems to fail on the curriculum front. The Australian curriculum is not remotely national, with NSW, Victoria and Western Australia implementing their own curriculum, which encompasses the Australian Curriculum. We also, connecting it to the lecture topic, need to consider the purpose of homework and what it is connected to.

Retrieved from on 20 November 2016

Alfie’s next point was interesting. He pointed out that we as adults have all been on the receiving end of homework and typically the vast majority of us as students neither liked nor wanted (and often did not complete) the assigned homework. Why, therefore, does homework persist? We do it because of rather than in spite of that history. I do wonder why so many parents continue to want their children to have homework. I had a few parents asking me about homework this term as I had not sent any home in the first two weeks and I responded with a message to all my parents via Class Dojo that I wouldn’t be assigning any homework other than some reading of interest and perhaps some SumDog, which they love. The students were relieved and I had a few messages from parents who were relieved as well.

So why does homework persist?

Homework has the capability to cause stress and frustration in students, frustration in parents; it can cause conflict between parents and their children, loss of time to other activities that a child would rather be doing and is passionate about.

So why does homework persist?

For those students who are interested and passionate about a topic,  homework or a unit around that topic has the potential to cast a pall over it and kill the passion and desire for learning (there is still mixed evidence for this either way from what I could see of a very quick search). When even those students who consider themselves academically inclined are pleased to have completed their homework and would rather be doing any one of a number of other things, often involving friends we need to ask ourselves why homework still exists. Alfie singled out reading logs as being particularly harmful. There is no better way, stated Alfie,  of killing a passion for reading than the enforced use of a reading log to monitor how much a child is reading. It becomes a thing  that must be done to get a reward from their teacher of a sticker or a stamp. If you are not sure about rewards and why they are potentially bad, read the previous article in this series.

Retrieved from on 21 November 2016

One of the alleged reasons that support homework is that it supposedly improves achievement. Before going into what Alfie had to say on the matter, the definition of achievement here is very important to this statement. In conversations that I have had, the achievement that is typically being referred to is testing scores. Testing scores are useful in many ways, but they are an indicator only of how well a student answered a specific set of questions at a particular moment in time. The familiar adage that goes something like what is tested is not important and what is important is not tested comes to mind here, though I do not doubt that there are those who would argue that testing is a valuable source of data.

Alife stated that no controlled study demonstrates any measure of improvement before a student reaches the age of fifteen and that even then, there is only a correlation not a causative relationship of statistical significance. Furthermore, at the High School level, the correlation between increased levels of homework and increased test scores is only a very modest correlation and explains very little of the growth.

An amusing example of how correlations can be meaningless and not explain a phenomenon. Retrieved from on 21 November 2016

Including the above graph is perhaps a tad facetious, but the website provides some laughs at the bizarre correlations that can be made between unrelated data. Alfie continued by commenting that the relationship between homework and improved achievement is only measurable in standardised tests which do not measure anything other than a student’s ability to answer that set of questions. An additional factor that needs to be considered is an interesting one, particularly in the case of internationally administered tests such as The Programme for International Student Assessment (PISA), The Progress in International Reading Literacy Study (PIRLS) and The Trends in International Mathematics and Science Study (TIMSS). These tests are initially developed in the English language and are then translated into the other languages required. What impact does this have on the terms chosen and the clarity of the question vis-a-vis the raw wording? How about the way a question is composed and understood culturally? How are questions transliterated (I hope that is the right term, please let me know if it is not) to ensure the meaning and underlying spirit/vibe/focus of the question are retained across cultures with different understandings and interpretations of different events and contexts?

These tests are initially developed in the English language and are then translated into the other languages required. What impact does this have on the terms chosen and the clarity of the question vis-a-vis the raw wording? How about the way a question is composed and understood culturally? How are questions transliterated (I hope that is the right term, please let me know if it is not) to ensure the meaning and underlying spirit/vibe/focus of the question are retained across cultures with different understandings and interpretations of different events and contexts? Alfie also told the audience that the modest correlation disappeared altogether when multi-variables are taken into account.

Alfie’s next point is an interesting one. He commented that cross-cultural tests, such as PISA and TIMMS show a negative correlation between tests and the amount of homework given. I want to believe this, and it feels instinctively true partly because I want to believe it (confirmation bias?) because a quick Google Scholar search returned a few articles that allege to have found positive relationships between the amount of homework given and general test scores (such as here, here and here), though that is based on reading the abstract only as they are all hidden behind pay-walls. The SmithsonianMag website has one very brief article which indicates that some homework has a positive impact on test scores.

Retrieved from on 21 November 2016

The above image was sourced from a well-written blog by Darren Kuropatwa (@Dkuropatwa) (though it appears to have been sourced from a PIRLS document) examining the value of homework based upon the Assessment Matters! publications by The Council of Ministers of Education Canada examining the data using The Pan-Canadian Assessment Program (PCAP), PISA,  PIRLS and TIMSS. The results, it seems, are still inconclusive when looking at this data.

Another reason that is often touted for the ongoing assignment of homework is that it builds character, self-discipline, organisation and other similar non-academic benefits, which I have recently seen referred to as soft-skills. Alfie remarked that no study has ever validated this belief, and to my mind, it is not even particularly logical, and even if it were, there are far better ways of teaching children those characteristics. I do not see it as logical due to the fact that the student has to do the homework, whether that is enforced at home by the parent/s or at school by the teacher and therefore no learning of such characteristics is going to occur. If it those type of soft-skills (are we really using that term?) that you want to teach your child/student, then look into having them take up a team sport to teach them getting along, collaboration and teamwork. Or a martial art for resilience, organisation and focus. Or have them complete minor age appropriate chores, involve them in conversations about organising the grocery list or the household budget.

The next argument for homework Alfie indicated he hears regularly is that it creates a window into the class which seems completely nonsensical to me. If you genuinely want to develop a relationship or connection between home and school, invite the parents to visit and help out with reading or maths groups, set up a class Twitter account for students to share what they are learning, or a class blog so they can publish their writing, art or audiovisual creations and reflections.

The final argument for assigning homework that Alfie spoke about was what he called the Beguti argument; better get used to it. This is, again, illogical and is used to justify a range of tools such as competitions, marks and group-work and, remarked Alfie, developmentally inappropriate. It seems silly to me, to use that argument. Because you will have to complete this task which is dreary and you will dislike it later in life, to prepare you for that dreariness, it will be inflicted upon you earlier than necessary. Children do not get better at dealing with negative things by having negative things happen to them at a younger age and using the beguti argument is, Alfie noted, akin to giving up the game.

Retrieved from on 21 November 2016

Parents are often sent home a homework letter that states something along the lines of your child will receive homework on x days or is required to complete the following tasks each week. This sends home the message that the individual needs of the child do not actually matter. Further to this, the audience was told that where schools advertise as teaching the whole child, but then assign homework, that they are paying the whole child lip service by assigning additional academic tasks.

We did hear some of the arguments against homework that Alfie has heard regularly. Parents opposing it on a value basis; that school work is for school and the family home is for family time, not school work. Additionally, homework exists, he has been told, because we do not trust children to occupy themselves without wasting time on Facebook, computer games etc and as an argument, implies that homework is merely busy work anyway. I find that an intriguing statement as I know that I, and every other adult I know, likes and needs to spend some time doing nothing, or, wasting time, to relax and unwind after being at work or to deal with stress.

Alfie qualified his dislike of homework by indicating that he can see some occasions when homework should be assigned, but that there should be some criteria for it:

  1. The homework will help students think more in-depth about the topic;
  2. The homework helps students become more excited about a topic;
  3. The task cannot be completed in school for some genuine reason other than we did not have time.

Assigning homework should be done after a class discussion with all students and the teacher agreeing that the homework is genuinely needed and that the default setting for homework should be no homework. Students who can complete a task and understand the task and how they completed it initially do not need to complete the homework, and those students who need the homework are likely unable to complete it and having to try and complete it will result in frustration, Googling the answer or a parent/guardian completing it for them.

What are your thoughts on homework in general and what I have written about Alfie’s lecture in particular? What are your thoughts on how this relates to cognitive load theory and the completion of basic tasks? I have reached out to Jon Bergmann for his thoughts on how Flipped Learning and the research which Alfie spoke about are related and I have also reached out to Alfie for feedback on both this article and the previous article and to check the accuracy of my interpretation of what he said.

I would very much like to hear your thoughts on this topic and I thank you for reading through both this and the previous article which are both rather lengthy.


Reflections on The Case Against Competition

Competition is a recipe for hostility. By definition, not everyone can win a contest. If one child wins, another cannot. This means that each child comes to regard others as obstacles to his or her own success. Forget fractions or home runs; this is the real lesson our children learn in a competitive environment.
-Alfie Kohn. Retrieved from on 19 November 2016

As I sit here at Adelaide Airport after an excellent few days at FlipCon Adelaide with Jon Bergmann, Ken Bauer and many other fine educators, I thought it prudent to begin organising my reflections on the evening spent listening to Alfie Kohn (@alfiekohn) speak. There was a lot to take in that evening, with two sessions, The Case against Competition and The Homework Myth and I was left with a strong sense of cognitive dissonance as a result. As you read this article and the next, I would like you to consider your own philosophy about punishment and rewards.

I had heard of Alfie Kohn during my Initial teacher Education (ITE) during my first-year Sociology of Education and my second-year Classroom management courses. All that I could remember from then was that Alfie promoted a laissez-faire approach to classroom management, however, I could not remember anything that was discussed vis-à-vis his views on competition. Alfie opened by remarking that the best teachers do not talk, they let the students do the talking. He also noted that the best teachers also do not test and find another way of fulfilling their requirement to provide a grade

Rewards are as bad as punishments
Alfie Kohn

This statement was, for me at least, a very unexpected and odd remark. Fortunately, Alfie spoke about some of the research behind the comment. He commented that in studies that have been undertaken about the impacts of rewards, that on average they reveal that providing rewards to participants for completing a task or achieving a goal results in less success than those participants who are asked to complete the same task knowing that there is no reward. This is particularly the case in studies where the goal has been to quit smoking or to lose weight with one set of participants being offered a reward and the other set receiving no additional intervention. Two studies, he continued, show that being praised for characteristics such as generosity, niceness, sharing etc. actually results in the participants becoming less generous, less kind, less willing to share; that the reward tends to undermine the characteristic being praised.

Alfie then asked us to pair up and develop a hypothesis for why being praised for social characteristics has been shown to be counter-productive. When many people moved into groups of three or four, he commented at the end that he did not mind that we had not actually followed his instructions worked in pairs for the task; however, he added that he hopes we are as tolerant of our own students doing the same thing, a remark which elicited a smattering of slightly nervous laughter.

He commented that he is often asked questions along the lines of “but what about acknowledging rather than rewarding?” and that he sees that there are many ways of distinguishing between the two. I do not actually have any notes indicating what he said about that topic and I cannot recall whether that was because he made that comment and then moved on or if it was because I simply did not write anything down. Personally, I do not see that acknowledging a student’s effort or results can be the same as rewarding them for the same. Something as simple as “I can see that you have been practicing x” or “I can tell that you have been working on y” is not, in my view, rewarding them, but merely acknowledging that you can see the effort, particularly if you then move on to providing feedback or whatever it is that you are doing.

Alfie then began to speak about punishments, or as they are often called; consequences. He remarked that the language does not change the fact that if a student feels like it is a punishment then it is a punishment. Forced social isolation sounds terrible, yet it is often referred to as detention or time-out and spoken as being a consequence of some action he said. Alfie continued by noting that to understand rewards, we need to understand punishments and that punishments (and rewards) can work. They can achieve temporary compliance in a specific context, but only at a great cost.

Alfie posits that when we tell a student “Do x or I will do y (or y will happen to you)” we are teaching them self-interest as it encourages students to think about the action as being necessary for the reward or avoidance of punishment. It also, in the case of punishments, enacted when the student is caught doing whatever the act is that is being frowned upon, which means that if they can avoid being caught in the act then they will do the act irrespective of the threat.  This creates distrust and fosters selfishness.

“The best way to ensure that something happens is to ban it.”

This desire to avoid being caught doing something deemed wrong and therefore avoid being punished stifles conversations about the kind of woman or man the child wants to grow up to be. The implication here is that the paradigm of how we teach children about right and wrong and about consequences, punishments and rewards is ineffective and detrimental to achieving what we want. However, this assumption depends on what the goal is. We should be asking students to consider what kind of person do I want to be?

I struggle with this. My (admittedly limited) experience tells me that students struggle to think about abstract things in a concrete way. That considering what their personality will be like when they are adults is a very difficult feat to achieve to with real sincerity. I have tried to talk with students about their goals for the future, some with very specific goals, and encourage them to consider how their learning now will impact their ability to achieve their goal and it is not something they are able to do.

Additionally, I struggle with the concept of removing punishments or consequences, which I get the impression is what Alfie is arguing for when I know that there are students with whom that approach will only exacerbate their behaviour, escalating it to be dangerous. I have friends who have been in classrooms and have had chairs thrown at them, or tables have been flipped and fights have broken out amongst students. I agree with the premise (Aristotle, Rhetoric) that all actions occur for a reason, yet how do you educate a violent child that violence is not the answer when, in their experience, violence gets them what they want? I have heard it said that if your lesson is engaging enough that you will not have class management problems. But with more experience and (a little) less naiveté, how can you deliver a lesson when you cannot engage the students because they are violent and disruptive or disengaged with school, or there is no support for education at home?

Retrieved from on 19 November 2016

Alfie posits that punishments and rewards are as manipulative as each other and are therefore merely two sides of the same coin. Rewards, he said, disrupt the vertical (teacher-student) and the horizontal (peer-peer) relationships and trust. Rewards reinforce power that the parent or teacher holds over the child or student. Alfie remarked that in classrooms where a reward is offered for a certain level of achievement, or grades that a student who needs and wants help is less disposed to ask for help in order to conceal the weakness. This on reflection seems rather counter-intuitive. If a reward is offered for achieving a certain grade and assistance was required to achieve the grade and therefore achieve the reward then why would you not make the request for help? Am I being too simplistic here in my understanding of the scenario? Perhaps this is the kind of context that Alfie spoke about, that if the reward is large enough then the result will be temporary compliance.

Retrieved from on 19 November 2016

One of the greatest predictors of learning, Alfie informed the audience, is being part of a caring community; with a sense of us rather than a sense of me. This links back to what was said about the way in which punishments create a sense of if I do not get caught and implies that a strong community will actually contribute to a reduction for the need for punishments. He also stated that just as punishments change the thinking to be ego-centric, to avoid punishment, so too do rewards change thinking to be ego-centric, to be given the reward.

An award is simply an artificially rarefied reward. Alfie posits that creating a competitive culture in the classroom tells students that others are an obstacle to their success and that their value or worth is only as good as the extent to which they defeat others. We were also told that competition makes students less communicative. The point that was made here was that why would a student communicate openly or trust others when it potentially creates a situation where their help will be used against them, fostering a culture where people are envious of winners and contemptuous of losers. Perhaps this is remedied by changing the system and removing grades and marks?

The impact on children of second-hand smoking. Retrieved from on 19 November 2016

In this day and age we are all aware and understand the dangers of second-hand smoke. Alfie said that “the impact of competitiveness is as bad as second-hand smoke in the psychological sense as it creates a situation where students cannot feel good about themselves in one context or another without others failing. I have not read enough of the research (read: I have not read any of the research) to have a grasp on the accuracy of this statement. My initial reaction was that it is a gross exaggeration and that that could not possibly be true, however, I have seen advertisements from the days when smoking was considered a good thing and I remember the arguments and the gradual social shift that took us from a society where smoking was acceptable to it being seen as a disgusting and smelly habit, with smokers shunned in many ways.

Alfie began to wrap up the session at this point. He reminded us that rewards and punishments focus on the end-result, not the underlying reason for the end-result and all behaviour happens for a reason.


Rewards undermined the interest of students in whatever it is that is being rewarded, the audience was told. I have not seen this for myself, however, I have seen how schooling can destroy a student’s interest in something through the edufication of a student interest .  My school is putting on a musical this year to celebrate the school’s sixtieth anniversary, calling it Dancing through the decades. Each Stage has been assigned two (non-consecutive) decades and will be performing a ten-minute block for each decade. Stage Three has been assigned the seventies and the current decade (are we really calling it the teens?) One of the songs chosen was quite popular when we first began but is now very much less so because they are sick of hearing the music and practicing singing.

The model is wrong, to the point where students are sometimes rewarded for a reward. Being rewarded for a reward, whether it be a car for good exam results in the case of those at the end of their schooling or something much smaller such as a toy, lollies or fast food for a perfect spelling test is wrong. We did not get explicit reasons for this, however, I feel that it is an amplification of the overall problem with rewards, that it undermines whatever is being strived for by limiting growth and putting a cap on excellence.

The topic of motivation arose, and Alfie reminded us of the typical understanding that there are two types of motivation, intrinsic and extrinsic,  and that, he alleges, research shows that the kind of motivation has a much more significant impact on the outcome than the amount of motivation. He went on to say that research indicates that as levels of extrinsic motivation increase, levels of intrinsic motivation shows a proportional decrease.


Rewarding someone for doing something actively damages a person’s interest in that thing. Alfie spoke about research that was done to study this using soft-drink. A new flavour was tested and those who drank it knowing that they would be rewarded showed more negative feelings about the product a few weeks later. In contrast to this, those who drank the product knowing they would not be receiving a reward demonstrated the same or slightly better feelings about the product afterwards.

Alfie posed a hypothetical: what is the alternative to rewards? Ultimately, it varies according to what the end goal is and no one alternative is a panacea to solve the issue. He indicated that it is his view that programs such as Positive Behaviour for Learning (PBL) do not achieve anything more than produce mindless obedience and that they do not result in more proficient learners. Giving students more say in the way that classrooms are run is one way of impacting attitudes and behaviours and I have heard of a system where a class runs their own court system (as outlined here) which apparently works quite well.

Alfie’s talk raised lots of questions and I feel like I still do not have answers for them. I did not have an opportunity to speak to him after the session as he, like all of us, needed a break to eat some dinner before the second session began. Reflecting on the ideas and what he spoke about, though, I can agree with much of what he spoke about. Where I begin to struggle is with implementing the changes that would be necessary to change my own practice.

What does this look like in a classroom? What does it look like in a school? Even if you negotiate the rules within a class and punishments for breaking them with the class and everyone is on board, is that generating the same results as rules arbitrarily set down and enforced by the teacher? Consequences are a part of life; touch a hot stove and you get burned, commit a crime and you face the consequences set down in law. If that is wrong, then the fundamental fabric of how our society functions is wrong and needs to be re-examined, which I do not see ever happening due to a large range of factors, not the least of which is tradition and a need for control (which in and of itself speaks to a lot of what Alfie was saying). If you do manage to create a model without punishments, how do you correct the behaviour which is socially unacceptable; swearing, derogatory language, violence, and disrespect of self and others; in students who see those things modelled as acceptable at home? How does this model fit with research from The Dunedin Project? This world-renowned longitudinal study has found that there is actually a genomic predisposition towards violence and extreme violence in some people, amongst other traits often regarded as socially unacceptable (Caspi, A. , McClay, J., Moffitt, T. E. , Mill, J.S., Martin, J. , Craig, I., Taylor, A., Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 2002, 297 (297), 851-854). . What impact does that have on this issue?

I often hear that secondary education cops it because things that happen or do not happen in primary school. I also often hear parents ask why I am trying a particular approach or idea when their child is off to high school and it will straight back to traditional classrooms, masses of homework etc. How does this concept fit within that constraint, the transition from primary to high school? There is a lot to consider here and I would very much appreciate hearing your comments on the topic, as well as any feedback on the article or your own interpretations of Alfie’s work in this area.

As always, thank you for reading.