For years, there has been a tension in primary and secondary education between testing and learning. Measuring students takes time away from learning, and stresses out students (and teachers and parents and administrators…). However, most primary and secondary education around the world is publicly funded through taxes, and the public has a right to know how students are doing. In the USA, anecdotal reports over the years of failing schools, in which students graduate high school lacking basic literacy and numeracy, periodically prompted calls for more high stakes testing and accountability resulting in, for one, the No Child Left Behind law.

This has snowballed to such an extent that we now have a crisis in over-testing. The Council of Great City Schools, which represents urban districts across the U.S., found that students take an average of 113 standardized tests between pre-K and senior year of high school.

Preparation for these tests has become a focal point for educators and administrators. The best way to improve systemwide scores on high stake tests is to improve teaching.. But while improving the results of teaching—through higher teacher pay, more teacher training and better funded classrooms—ought to be the goal, it is difficult to achieve at scale. The easiest way to improve test scores systemwide is to dial up the average amount of homework per student. A study conducted by researchers at Indiana University, University of Virginia, and University of Macau and published in The High School Journal found: “There is a consistently positive significant relationship between homework and performance on standardized exams.”

Test Prep Cram Schools

The result is that public school classrooms increasingly look like test prep cram schools. The flipped classroom model has given this creeping homework inflation pedagogical cover. But rather than working problems in class and watching video lessons at night, as in a traditional flipped classroom, too many public school students today are working problems in class and then working more problems at home.

In a conversation with ASU President Michael Crow at the Aspen Ideas Festival a few years ago, I discussed this very idea: “Instead of teaching an intuitive approach to math, we’re turning schools into test prep centers. Our society instructs teachers to prepare kids for rapid-fire multiple-choice questions. Big surprise that the kid hates math, then.”

That said, students do need practice, and a lot of it. Musicians practice. Athletes practice. Those who practice more tend to improve more than those who don’t. I spent years in the test prep industry, working for Kaplan Inc. I saw a lot of student data and the results were clear: the more students practiced by doing extra problems, the greater their proficiency gains.

What we need is to strike the right balance. In recent years, the pendulum has swung to excessive drilling. Which isn’t much more fun in school than it is at the dentist.

Personalized learning can help. If we can differentiate homework for each student, we can employ techniques that weren’t previously available:

  • Don’t assign problems that will be so difficult for particular students that they will very likely fail, and may become discouraged
  • Don’t assign problems that are so easy for particular students that they have no assessment or instructional value and are just wasting those students’ time

These techniques have been brewing for some time now. They are commonplace in adaptive learning systems that deliver homework on devices. And, increasingly, teachers are finding ways to do this manually for paper-based homework in their own classes. It’s time-consuming and difficult to do this, and my company Bakpax hopes to help automate this. (More on this another day.)

Putting Content in Context

Another technique I’d like to see is what I call “content in context.” The idea is to personalize content based on what might be more interesting, and hence engaging, to particular students. Experienced educators often give different kinds of examples in class, that are particularly relevant to a subset of students—e.g., the example of a football field in an high school geometry class or the example of a psychological study in a college level statistics class. One must be careful to keep these contextual examples balanced. Using sports examples predominantly is unfair to students indifferent to sports.

While we don’t know how beneficial the use of content in context is, it appears that not using context is harmful. Gregory Palardy, Assistant Professor of Education at the University of California Riverside, found in a study that defined context as the “composition of the student body, classroom structures, and resources” that context was the greatest contributor to the achievement gap among Black and Hispanic children.

When you step into an experienced educator’s classroom, she is often integrating real world examples to build connections with the material. Consider the ELA teacher who uses a student government election to teach persuasive writing. Or the veteran math teacher who demonstrates that the recommendations in Netflix are based on data science. Or the science teacher who explains light theory by talking about how this impacts set design for musicians on massive world tours.

These educators apply context to make their assignments are more effective. They use the context of their individual classroom setting—where their students come from, prior knowledge levels, and cultural norms—to develop a deeper level of comprehension by putting content in that context. Outstanding teachers attempt to employ a variety of contextual hooks.

The problem is that, today, the onus is on the teacher to do it all.

Today it is impractical for teachers to assign personalize homework based on context. To be sure, many do, especially in soft side disciplines. Hard side disciplines like math and science could come alive for students given their fertile ground for interesting examples and practice problems. (I once blew my own mind in college when I was studying the concept of escape velocity, and I calculated that Mars’ moon Phobos has enough gravity that I would stick to its surface, but little enough that I would achieve escape velocity if I just jumped really hard.) But it’s difficult to do so systematically, even in the soft side, without large volumes of differentiated content.

Imagine a New Line of Textbooks

This is where textbook publishers could play an incredible new role, if they so choose. Imagine a new line of textbooks that offered multiple kinds of examples for each new concept, each employing it in a different real world application. Imagine a textbook where practice problems tended to be stories or puzzles.

Textbooks publishers are challenged today by free web content, rental markets, and digital products. It’s time to innovate! Make the paper textbook better than it’s ever been, with more content in the teachers’ editions for the teachers to pick and choose which examples they want to use to illustrate the lesson and which questions they want to assign based on what will be most engaging and relevant to the class.

Publishers should think about diversifying assignments to better align to students’ range of interests and circumstances, and should concentrate on making content meaningful and relevant. Just think about the outcomes if students could self-select practice assignments that match their interests and feel relevant to the world around them. The more a student becomes engaged in homework, the more practice questions they will complete. And the more practice that is both completed and understood on a deeper level, the closer we’ll be to meeting the goal of homework in the first place.

We’d love to hear from teachers. How do you use context in providing more meaningful content? How do you make homework more effective?

Photo from The Sopranos