Differentiated Homework

In my last post on homework I had an interesting question posed by a reader:

samjshah wrote:

…would you ever consider assigning different amounts of problems based on how well/poorly students did on the last assessment? Like saying, if you got a 90-100% on the last test, you should do homework I, if you got a 75-89% on the last test, you should do homework II, and if you got less than a 75%, you should do homework III.

And the homework might be something like this:
Homework I: 5 mid-level problems checking understanding]
Homework II: 5 beginner and 5 mid-level problems
Homework III: 5 really easy, 5 beginner, 5 mid-level problems

In other words, some sort of differentiated homework scheme?

As I started to write an answer in the comment section I realized that it really deserved a post of its own. This is an interesting idea, and I confess I have never done anything like this, nor even thought of anything this prescribed. I definitely see some positive aspects to this concept, but some negatives as well. Let me start with the positive ones.

First of all, it recognizes that students are not widgets and that they have different needs. I like that the different levels of homeworks are not merely based upon having differing numbers of problems, but also on the level of the problem. This definitely denotes a recognition that students who are struggling need basic practice as well as continued work on more conceptual problems. A question I have is are the problems assigned, or do the students have a choice within the parameters given?

Here are my concerns, and I realized I had more of these as I continued to think about it. First of all, and on the most practical level, this seems like a nightmare to organize and keep track of. Choosing the problems can be difficult, although somewhat mitigated if you have a textbook with A, B , and C level problems. Checking the problems, and keeping straight who has how many (and what type) of problems due could be difficult – particularly in a class of 37, which were my class sizes for the first few years I taught.

I find that my time is more well spent on planning engaging classes and thinking of ways for my students to understand the material than it is spending a lot of time checking through homeworks. If I assign homework I feel obliged to honor that in some way, but the more complicated I make it (and I have come up with some bizarre homework schemes over the years) the more superficial my responses are.

Another concern is that the grade on the test seems to suggest that the main (or only) reason for a particular grade is based on the amount of homework done. There are a multitude of reasons that students will do well on tests or not besides homework. Unless the questions on the test are identical to homework (a thought I will need to post on later) then doing the homework may not necessarily result in better test scores. And what about the student who will inevitably feel ‘punished’ by having to do the extra problems every night all year because no matter what they do they struggle on the tests?

Using the format that I created, with the optional problems and 3 questions, I did have students abuse the system and do no homework in the beginning. Their test grades suffered accordingly and this became a fruitful discussion between me and the student – with them coming to the realization that they needed to do more homework rather than me dictating it. My experience previous to this was that the more I told students they needed to do the homework, the more likely I would get blowback from them and even if they ended up doing it, they did it fairly mindlessly. They might finish the ten problems I assigned, but they made no connection between the problems. By allowing them to have choice about the problems they did (and by the end of the year a majority of students not only did the recommended problems but the suggested ones as well) and making the homework that was actually turned in reflect what they learned I found they became more conscientious – and a side effect was that the students felt I was on their side.

In conclusion, the idea behind the differentiated homework idea is a good one. It is important to recognize that students are different and what works for one student will not necessarily work the same way for another. We are somewhat constrained by logistics in terms of differentiating too much, but the more we can do it, the better for our students in the long run. The one other key thing I bring to this discussion is responsibility. I am of the mind that we give the students as much of the responsibility for their own learning as we possibly can, and then help them make the connections between the choices they make when exercising that responsibility and the consequences that naturally follow.