Elaborating a Framework for Contextualized Problem-Based Instruction

In classrooms across the United States, a quiet revolution has been reshaping how mathematics is taught. For decades, the standard method involved a teacher demonstrating a procedure on a board, followed by students practicing similar problems until the steps became automatic. Today, a growing movement encourages a different path: presenting students with a novel, real-world problem first, asking them to struggle with the solution before any formal rules are introduced. This approach, known as problem-based instruction, relies heavily on context. Instead of abstract numbers floating in a void, students might calculate the cost of a road trip, design a garden, or analyze sports statistics. The hope is that by anchoring math in the familiar world outside the classroom, students will find the subject more meaningful, accessible, and useful. However, while the philosophy is clear, the practical mechanics of how to sequence these lessons remain murky. Educators and designers often lack a clear map for navigating the journey from a messy, real-life situation to a precise mathematical answer. They know they want students to connect the two, but they do not always have a shared language to describe the specific steps required to build that bridge.

This is the gap that researchers Luke T. Reinke and Xin Liu set out to fill. Working with three of the most widely used middle school mathematics curricula in the country, they analyzed thousands of questions and tasks to understand exactly what happens when students engage with contextual problems. Their goal was not to invent a new theory, but to create a practical framework—a set of categories—that could describe the different types of thinking students are asked to do. By breaking down the complex process of solving a real-world problem into distinct, observable actions, the researchers developed a tool that allows curriculum designers and teachers to see the structure of their lessons with new clarity. They wanted to know: what kinds of questions are actually being asked, and in what order?

The researchers examined materials from three major programs: Illustrative Mathematics (also known as Open Up Resources), Eureka Math, and iReady. They focused specifically on seventh-grade units, selecting five lessons from each of the five major mathematical strands to ensure a broad view. They did not just read the lessons; they dissected them, segmenting every task, question, and instruction into small pieces. Each piece was then coded according to a specific category of activity. Through this rigorous process, they refined an existing framework, adding new categories and clarifying old ones until they had a robust system that could reliably describe the flow of a mathematics lesson.

The result is a detailed map of six primary types of activity that occur within contextual problem-solving. The journey often begins with a focus on the context itself, where students are asked to activate their prior knowledge about the real-world situation, such as discussing their experiences with weaving or road trips. This is followed by a crucial step where students must formulate the problem, identifying what is actually being asked, what information is relevant, and what assumptions they need to make. In some lessons, this step is explicit, asking students to decide what data to collect; in others, the problem is already well-defined, and the task is simply to identify the key numbers.

Once the problem is framed, students move to producing a model. This is the moment where the real world is translated into a mathematical representation. Students might draw a diagram, create a table, or sketch a graph to represent the relationships they see. The researchers found that these models can be informal, like a drawing of spoons and cups, or more structured, like a pre-formal ratio table. After creating this representation, students use the model to answer a specific question about the context, such as determining how far a family will travel in a given time. This step tests whether the model actually works to solve the original problem.

However, the learning does not stop at finding an answer. The framework identifies a vital phase where the focus shifts entirely to the model itself. Here, students are asked to work within the mathematical structure, perhaps translating a drawing into an equation or calculating a specific value like a rate of change. This is where the math becomes abstract, detached momentarily from the story to explore its internal logic. The final, and perhaps most important, step is focusing on the correspondence. In this phase, students are asked to look back at the connection between their abstract model and the real-world situation. They must explain what a specific number or graph feature actually means in the context of the story, or evaluate whether their answer makes sense in the real world. For instance, they might be asked if it is reasonable to assume a car drives at a constant speed for ten hours straight, forcing them to critique the limitations of their own model.

The researchers also discovered that these activities rarely happen in isolation. They frequently appear in specific sequences that create powerful learning opportunities. One common pattern involves students producing an informal model, using it to solve a problem, and then being guided to translate that model into a more formal mathematical representation. This sequence mirrors a concept known as vertical mathematization, where students move from their own intuitive strategies to standard mathematical conventions. Another frequent sequence asks students to identify a mathematical parameter, such as a constant rate, and then immediately interpret what that number means in the story. This back-and-forth helps students build a deep understanding that links abstract symbols to concrete reality, preventing the math from becoming a meaningless set of rules.

The study also highlighted what is missing. While the curricula were rich in many areas, the researchers noted a scarcity of tasks that ask students to generate their own questions based on a model. In the high school materials they had analyzed previously, students were sometimes asked to pose new questions that could be answered by a given equation. In the middle school materials, this type of creative problem-posing was largely absent. This omission is significant because the ability to ask questions is a core part of mathematical thinking, yet it is often overlooked in standard lesson plans.

The framework developed by Reinke and Liu offers more than just a way to categorize questions; it provides a menu of options for those who design education. By understanding the distinct types of activities available, designers can intentionally sequence lessons to ensure students experience the full cycle of mathematical modeling. They can check if a lesson starts with the right kind of context, moves through the necessary steps of building and using a model, and ends with a reflection on what the math actually means. The framework also helps researchers and teachers analyze how instruction is enacted in the classroom, allowing them to see if the intended sequence is being followed or if opportunities for deep understanding are being skipped.

Ultimately, this work suggests that effective problem-based instruction is not a single method but a carefully orchestrated series of different types of thinking. It requires moving students from the familiar world into the abstract realm of mathematics and then back again, ensuring that the connection between the two remains strong. By providing a clear language to describe these movements, the researchers have given educators a tool to build lessons that do more than just teach procedures. They offer a path to help students see mathematics not as a collection of isolated facts, but as a powerful way to understand and interact with the world around them. The framework does not claim to have solved every challenge in mathematics education, but it provides a reliable foundation for designing instruction that is both principled and practical, ensuring that the journey from a real-world problem to a mathematical solution is clear, coherent, and deeply understood.