Homework: What is it good for?

Part of the problem with my particular leadership style is that I am very reluctant to tell other professional educators what to do.  I have my philosophy and my classroom policies, and they are free to have theirs.  A big part of my reasoning for this is recognizing that what works for one teacher might not work for another.  We each have a unique classroom–the students, the space, the resources, and our own personalities–and we need to make decisions based on that unique situation.

Homework is a prime example.  I don’t give very much homework.  This is not for the reasons that you might think, such as valuing my students’ time out of class or questioning the importance of this work.  It’s because I don’t trust my students.  I am not naïve enough to believe that when they take work out of my class the work that they return is solely the product of their effort.  Any veteran teacher can list a dozen different factors that contribute to the quality of homework, other than the skills of the student.  As I put it when I am explaining my policies to my students, “I simply don’t know who completes the homework that I assign or what resources he used to do it.”

But, I don’t fault other teachers who choose to assign, check, track, and grade homework.  If they see value in it, and it helps their students succeed, they should keep doing it.

Now, however, comes a new study that seems to reinforce the idea that homework has value.  The authors point to the statistical relationship between the time (self-reported) that students spend on math and science homework and their performance on standardized tests.  It should come as no surprise that students who practiced mindless tasks nightly also did well on mindless tasks administered over hours at the end of the school year.

What is surprising, and easily overlooked, is that when the authors looked at the relationship between homework and grades in class, they couldn’t find one.  As Alfie Kohn points out, this is amazing when you consider that the same teachers assigning the homework are the ones assigning the grades.  Wouldn’t you assume that, since most teachers who assign homework make it count as a portion of student grades, this relationship would be stronger than the one connecting homework time with test scores?

The answer is puzzling, but the data don’t lie.  Taking a “big picture” view of homework effectiveness seems to reveal that it might not be worth the time and effort that both students and teachers are putting into it.

What’s your stance on the value of homework?

photo credit: Cayusa via photopin cc


Grading Metaphanalogies [Guest Post]

My wise and well-spoken colleague, Erica Speaks, keeps the conversation going this week with another guest post, while I work on Kenan stuff.  Yet again, she makes a strong case:

You know, someone pointed out to me that anyone who reads our online debates from afar would think Paul and I don’t get along at all and must spend every opportunity finding something to ferociously debate. Well, that last part may not be far off, but overall, we actually agree on most things – pedagogically and otherwise. Honest.

We often discuss, debate, and think of things in terms of metaphors and analogies. Paul coined the term “metaphanalogies”. I like it – it’s how things make sense to me. So, when there’s a topic like grading where we have gone round and round about it, you can imagine we’ve used several. Well, he’s offered several, and I’ve shot them down. I’ll explain.

SPORTS: “The scoreboard reflects the game, not the practice.” Paul has pointed out that we judge players based on their performance in a game, not how they do in practice.

My problem with this metaphor is that in it, we the teachers are not fans, but coaches. Coaches of course judge players on their practices. In fact, players are denied playing time or even cut from teams. In the pros, players are fined or suspended from games, based on things that may happen in practice. Imagine, “If you don’t do your homework all week, don’t even bother showing up to third period.”

Also, in sports kids try out and are placed in classes or teams by their ability levels. (I don’t teach in leveled classrooms, but if you do, perhaps this aligns more for you.) Understand that I’m not arguing public school classes should have these things in common with sports. I believe in equal access to learning. And luckily, no one has to turn a cartwheel or throw a perfect spiral for success in life. They do, however, probably have to get through middle school. Again, more reasons the sports metaphanalogy doesn’t really work for me.

MEDICAL: Paul discusses here adjusting a patient’s temperature to encourage arriving to the appointment on time. This came up again in our discussion where he was “air quoting” me here.

As for temperature, I have a problem with comparing grades to something which student have no control over and can exert no effort to change. While I agree a grade is a measurement, I don’t agree that it is only a measurement. Height is only a measurement. If only people over 5’8” were allowed in college, I’d view this as a prelude to pessimistic resignation for many students.

So, what “metaphanalogy” correctly reflects grading? Throughout our discussions, I’ve really wrestled with this question. I knew when I answered it, I’d be able to get my head around the work that grades should reflect and the message they should send.

I realize it looks like I’m writing to a recurring theme, but I’ve determined that weight is not a bad analogy to grades. When this clicked for me, I had to concede several things to Paul’s arguments, but I think he owes me a few in return as well. 🙂

Even when one carefully follows a diet and exercises, the pounds on the scale do not always respond. Yes, you hope that effort yields results. Logic even suggests it; however, it cannot be guaranteed. For example, some people have conditions, like diabetes or a slow metabolism, that (much like a learning disability or lower IQ) make it an unequal playing field.

So, Paul, I hereby publicly acknowledge that the practice, the effort, should not be expressly configured in results. (Would that it were – many a time I have been faithful on my diet but not a pound has left me.) We don’t just read the scale differently because we put in the effort.

However, my resistance to discount effort lies in having experienced how hard it is to keep trying when results don’t come as easily as they might to others. (And yes, I am admitting that in this analogy, I am the equivalent of the “slow learner”.) I worry about that frustration point with our students. I fear those who need encouragement the most will simply give up because they will never measure up.

This issue is intensified by the reality of a fixed point of mastery for all. You could take me down to the skeleton, and I’d still be larger around the middle than the eighty-pound four-and-a-half foot teacher at our school. The level of mastery (in standardized tests, for a passing report card grade, or for the cutoff for honor roll – wherever one draws a line) is the equivalent of saying, “To be healthy, everyone will weigh 125 pounds, regardless of height, gender, age, etc.” Some could eat junk all day, smoke, use illicit substances…and still make that mark. Others could do everything ‘right’…and never make it.

“Grades shouldn’t be a motivating factor. Find other ways to motivate them.” Paul says this too, but what “other ways” should teachers try? Rewards, stickers and prizes? (Alfie Kohn’s Punished by Rewards anyone?) Research aside, extrinsic rewards don’t seem to be an organic way to foster what we want, anyway. Don’t give me stickers for completing a workout. Value the change made, the weight lost, ie: the growth in learning. However, growth-alone models are flawed. Just like the students who score the highest on our standardized tests have a hard time showing growth, that 80-pound teacher would have a hard time in a model of “whoever loses the most weight wins.”

This “metaphanalogy” works for me for one last reason. While we can boil them down to a convenient, single number (the grade, the pounds) this can be dangerous, even though it is so simple to do. (Or maybe because it is.) Weight is an indicator of health, but not the “be all end all”. There are many other numbers to consider – and other factors not expressed in neat, mathematical digits. We have SAT scores and standardized test scores that are supposed to be cut-and-dry measurements of mastery. Should none of the averages in a class value the journey? And if not, where should it be valued and counted?

Paul & I agree that sweeping changes are needed, and that it will involve a re-education of all stakeholders. For me, we could start with: change the notion that all mastery should look the same. Perhaps a child that makes one year’s growth on the standardized test would be marked as “passing” – regardless of where s/he is in comparison to peers? Should a child with a 70 IQ perhaps get an A if s/he earns 70% mastery?


Grading Homework: More Food for Thought

My recent post entitled “Am I Not Being Clear?” sparked more conversation than any other single piece that I posted here in several years.  The idea of grading reform and of motivating students without grades really struck a nerve with some.  When I challenged teachers who struggle to understand why I don’t include homework in my students’ grades to test it out themselves, I wrote,

“Based on my personal experience and anecdotal stories from colleagues,less than 15% of students fall into the third category [of students who only do homework when it affects their grades].”

In response, a regular Scripted Spontaneity reader and commenter (who uses the moniker “Last Teacher Standing”) challenged these numbers.  She conducted the experiment that I suggested and reported back with her results.  She wrote,

“This means that when it ‘counted’, 90% had their homework. When it didn’t ‘count’, 48% completed it.”

My first response was “Wow”.  Not only did she push back in that meaningful way that I so value in PLN, but she also found results that were starkly different from what I and others have seen.

As I imagine most people would feel, my first reaction was to defend myself.  I wanted to reply with a comment about how her results didn’t really change anything.  Or, that it was a fluke that wouldn’t hold up to further observation.  I wanted to prove that she was wrong and I was right.

Thankfully, my better judgment took over, and I look closely at her data.  I started to think about the differences between my classroom and hers.  I reflected on the conclusions and explanations that one could draw from a side-by-side comparison of these two groups of students (hers and mine).  Here’s what I came up with:

  • Students differ considerably in their performance between classes, teachers, schools, states, and countries*.  The results of one group (or even several groups) should never be assumed to represent any significantly larger group.  This is the arrogance of statistical sampling, and I fell prey to it more readily than I should have.
  • The motivation that grades provide for some students to complete their homework is a false one.  Extrinsic motivation is short-lived (according to Dan Pink, it only lasts as long as the reward is there).  If we truly value a strong work ethic and believe that our students need to learn to work hard because it will reap rewards later in life, than we must find other, more authentic ways to reinforce this skill.
  • The disfunction in ubiquitous ABCDF grading systems is profound.  We have so completely obscured the true meaning of a letter grade (at the end of a term) that it is nearly impossible to know from this one measure whether a student has learned the state-mandated curriculum.  It’s not just that teachers include other information in that grade; it’s that parents often expect it to include theses “extras”.  They expect that an A represents how hard their child worked that marking period, how well she behaved, how much homework she completed, how well she studied, how polite she was to the teacher, and–oh, yeah–how much she learned.  It’s a system that feeds back into itself over and over again.  Teachers say that they need to grade homework so that parents will see zeroes and make the children do the work.  Parents say that they make their children do the work because the teacher gave their child a zero.
  • Many teachers, through their classroom procedures and policies, create a culture of high expectations that results in extremely high rates of homework completion.  While you could argue that this positive outcome comes at a cost in terms of conflating work behaviors with content mastery, there is no debating its effectiveness in many cases.

So, what’s the solution?  Clearly, some teachers can’t (and shouldn’t!) just go cold turkey and stop including homework in students’ average.  Reporting systems must first be in place, along with rewards and consequences at the outset, to provide information about homework (and classwork/participation/behavior) habits to parents and students in meaningful and informative ways before removing the grade-based motivation.  We must make this transition, however, if a letter is ever going to be a meaningful measure of student curriculum mastery

That’s the message I took away from Last Teacher Standing’s data.  If the number of students who are only motivated by grades (directly or through their parents) is even higher than I thought than the problem is even more serious.  We must find a different way to get those grade-motivated students (15% or 42%) to do their practice work, because someday in their lives these students won’t be earning a grade for completing their TPS Reports:




*Note: I happen to know that Last Teacher Standing‘s students and my own are very similar, so this point is less relevant to this situation.