“It’s not working.”
A group of my 4th graders stared at their Cubelets robot, pieces scattered on the table. They’d connected a battery cube, a drive cube, and a distance sensor. They assumed their robot would move forward when something approached it. But nothing was happening.
One of my students rotated the cubes and tried again. Another suggested adding a different sensor. A third student added that the robot worked before, so it must be broken now.
Eventually, one of them looked up and asked me the question that teachers hear all the time.
“Can you help us?”
I hesitated. In many classrooms, this is where the teacher steps in to solve the issue. After a quick fix, the robot begins to move, and the class carries on. But I knew that when it comes to robotics and coding challenges, it is all about the process. Students need to learn to go through steps and try things multiple times. Real learning happens before the solution appears.
So instead of fixing the robot, I asked a simple question. “What do you think might be happening?”
My students leaned back into the problem by testing ideas, adjusting pieces, and debating what could be wrong.
The Importance of Struggle in STEM Education
Lessons on robotics and coding often highlight an important fact about STEM learning. Technology rarely works the first time. Sensors could be facing the wrong way. Connections could fail. Programs may behave unexpectedly. These moments may feel frustrating for students, but meaningful learning often starts here.
Unfortunately, in many lessons, teachers design and script the moments of problem solving. Instructional documents remove the burden of uncertainty, and instead guide students step-by-step toward success on their first attempt.
When problem solving disappears from STEM lessons, something important is lost. Students don’t get to experience the challenge that computer scientists and engineers face in the real world when working on new projects. A programmer will write code, test it, and anticipate failures before it works. An engineer rarely solves a problem on the first attempt. The process is fundamental to learning.
So why would teachers eliminate this part of the learning? They may do so without consciously realizing it. Teaching in a timed environment pushes educators toward step-by-step instruction to be as efficient as possible. When the pace of an activity slows, or students experience frustrating moments, many teachers find it easiest to intervene and guide the group forward.
Some common ways that this happens include:
- Teachers give out so much information during instruction that students don’t have the opportunity to wonder or explore.
- A student who makes an incorrect choice is immediately corrected before they can analyze what happened.
- Tasks are structured so that every student arrives at the same answer in the same way.
This eliminates opportunities for deeper thinking. Students become concerned only with doing the task “right,” rather than exploring possibilities and testing ideas.
When problem solving disappears from STEM lessons, something important is lost.
Designing Lessons That Encourage Productive Struggle
Allowing students to struggle productively does not mean leaving them without support. Instead, teachers can design learning environments where exploration and persistence are expected parts of the process. A few instructional shifts can help teachers create these opportunities.
Start with the challenge, not the instructions.
Rather than explaining every step at the beginning of a lesson, present students with a goal. In a robotics lesson, for example, students might be asked to design a robot that reacts to light or moves toward an object. Once students understand the challenge, allow them time to experiment before offering guidance.
Normalize iteration and revision.
In engineering and coding, first attempts rarely succeed. When teachers emphasize that failure is part of the design process, students begin to view mistakes differently. Instead of seeing errors as setbacks, they recognize them as opportunities to gather information and try again.
Ask questions instead of giving answers.
When students ask for help, responding with questions can encourage deeper thinking. Asking “What have you tried so far?” or “What do you notice about how the robot is responding?” helps students analyze their work and generate new ideas.
Build in time for troubleshooting.
Many STEM lessons struggle because schedules leave no room for debugging. Allowing students time to test ideas, make mistakes, and revise their designs ensures that problem solving becomes part of the learning experience rather than an interruption.
These strategies shift the teacher’s role from director to coach. Rather than guiding every step, the teacher supports students as they explore and refine their thinking.
Teachers can design learning environments where exploration and persistence are expected parts of the process.
When Struggle Is Allowed
In my doctoral research on how elementary teachers sustain student engagement during STEM activities, teachers consistently described exploration and persistence as essential parts of meaningful learning. When students are given time to test ideas, revise their thinking, and solve problems independently, engagement tends to increase rather than decline. And when teachers leave room for productive struggle, students gain more than just technical skills. They learn curiosity, resilience, and the willingness to tinker.
None of these skills develop by simply following a set of directions. They grow when students grapple with ideas over time and discover solutions through experimentation.
The Moment the Robot Finally Moved
At last, my students working with the Cubelets robot figured out the problem. The distance sensor was misaligned. After they turned the cube, the robot immediately rolled forward.
They cheered.
The team began modifying materials again, adjusting cubes, reconfiguring, and resetting as they explored new possibilities.
More than the thrill of their success, they relished the challenge of figuring it out.
At that moment, seeing the robot finally move was exciting. But the real learning had already happened.