Why Students Learn More When They Build Instead of Memorise

 

Here is a question that sounds easy but isn't. Which of these two students understands how a circuit works?

The first has memorised the definitions of current, voltage, and resistance, and can write out Ohm's law. The second has never been asked to recite a definition, but has spent two weeks designing, wiring, and fixing a circuit that makes a small fan spin.

Most teachers would say the second student understands it better, and the research tends to agree. But the surprising thing is how often the first student feels like they have learned more.

What 225 Studies Found

In 2014, Scott Freeman and colleagues published a landmark meta-analysis in PNAS. They compared traditional lecturing with active learning in undergraduate science, technology, engineering, and mathematics courses, pooling 225 studies.

The results were striking:

• Exam and concept-test scores were higher with active learning, by about 0.47 of a standard deviation, which the authors estimate at roughly 6% on average exam scores.
• Failure rates averaged 21.8% under active learning, compared with 33.8% under traditional lecturing, meaning students in lecture-only courses were about 55% more likely to fail.
• More was better. Effects tended to be strongest where a larger share of class time was spent on active learning.
• It worked across class sizes, and the gains were larger on concept inventories than on course exams, suggesting active learning deepens conceptual understanding.

Active learning in these studies covered a range of approaches, including problem-solving, group work, and workshop-style designs.

The Paradox: Learning More, Feeling Like You Learned Less

A 2019 Harvard physics study by Louis Deslauriers and colleagues compared lectures and active learning for the same content. Students scored higher after the active sessions, but felt they had learned more from the polished lecture.

This explains something important. Memorising and listening to a clear explanation feels smooth and productive. Building something is effortful, messy, and full of dead ends, and that effort can feel like not learning. As we saw in an earlier post on trial and error, learners often misjudge which methods help them most. The result is that students, parents, and sometimes teachers prefer the approach that feels easier, even when the harder one teaches more.

Why Building Works: The ICAP Framework

Educational psychologists Michelene Chi and Ruth Wylie offer a useful way to understand why. Their ICAP framework sorts student engagement into four levels, predicted to produce increasing amounts of learning:

1. Passive: listening, reading, watching.
2. Active: doing something physical or selective with the material, such as highlighting, copying, or repeating.
3. Constructive: generating something new beyond what was given, such as explaining in your own words, designing, or creating.
4. Interactive: building ideas together with others, through dialogue and collaboration.

The framework predicts that learning rises across the levels: passive is lowest, interactive highest.

In the classroom, the same topic can be taught at every level:

• Passive: listening to the teacher explain circuits.
• Active: copying the circuit diagram from the board, or memorising the definitions.
• Constructive: designing a circuit to solve a new problem, such as spinning a fan or lighting an alarm.
• Interactive: a team designs, tests, critiques, and improves each other's circuits.

Building naturally moves students up the ladder. They aren't just receiving or repeating information. They are producing something, testing it against reality, and often discussing it with others.

A Fair Word for Memorisation

None of this means memory doesn't matter. Learners need a base of facts and vocabulary to think with, and retrieval practice, recalling information from memory, is one of the best-supported ways to make learning stick. The problem isn't memorisation. It is memorisation alone.

A student who has memorised a definition but never used it holds knowledge that is brittle and hard to apply. A student who has used the idea to build something holds knowledge that is connected, tested, and usable. The most effective classrooms combine both: enough direct instruction and practice to build a foundation, and enough building to make it real.

What the Evidence Doesn't Show

It's important to be honest about the limits:

• The Freeman meta-analysis covered university STEM courses, not school-age children, and many of the studies were not randomised. The pattern is consistent with research at other levels, but the numbers shouldn't be transferred directly.
• "Active" isn't the same as effective. Following a fixed kit recipe step by step is more active than listening, but it isn't necessarily constructive. The benefits come when students generate, decide, and test, not just follow.
• Building takes time. Projects can use more class time than lectures, and need careful design to make sure core concepts are covered.
• Students may resist at first, precisely because it feels harder, so teachers need to explain why it works.

What This Means for Indian Classrooms

NEP 2020 calls for a shift from rote learning toward experiential, competency-based education, and CBSE's Kaushal Bodh and Composite Skill Labs bring hands-on Robotics, Coding, AI, and IoT into the regular school day. The research suggests these are not merely nice additions. They move students up the ICAP ladder, from receiving information to constructing and applying it.

How STEM-Xpert Puts Building at the Centre, From Grade 1 to 12

Everything we do is designed so that students construct, test, and share, not only listen and recall:

• Robotics Labs and AIoT Labs, where students design, build, and program real systems, and have to make them work.
• STEM Innovation Kit, India's first IoT-enabled STEM kit, and Crinnoboard, our own AI-IoT development board, with which students build robots, home automation systems, weather stations, and more.
• Tinkering Labs and Innovation Spaces, where students define their own problems and design their own solutions, the constructive and interactive end of the ladder.
• Composite Skill Labs and Kaushal Bodh centres, aligned with CBSE's CTAI framework, embedding building in the regular school curriculum rather than leaving it to extra-curriculars.
• STEM Academy, which combines digital and live learning, with projects and personalised support, so foundations and building reinforce each other.
• STEM Facilitator Course, preparing teachers to guide students from following instructions to constructing their own solutions, and to explain to students and parents why the effort is worthwhile.

The Bigger Picture

Students learn more when they build because building asks more of them: to generate, to decide, to test, to discuss, and to fix. The evidence from large reviews of STEM teaching, from classroom experiments, and from cognitive science all points the same way. The most important lesson may be the hardest one for schools and families to accept, which is that effort that feels harder can teach more. Schools that help students and parents understand this, and then give students real things to build, are giving them the kind of knowledge that lasts.

If your school wants students to learn by building, through Robotics Labs, AIoT Labs, Tinkering Labs, the STEM Innovation Kit, the Crinnoboard, or Composite Skill Labs, get in touch with STEM-Xpert to see how our programs can be set up for your students, from Grade 1 to 12.

FAQs

1. Does building really help students learn more than memorising? Large reviews of STEM teaching suggest so. A meta-analysis of 225 studies found higher exam scores and lower failure rates with active learning than lecturing, and psychology research ranks constructive, interactive activity above passive receiving.

2. Why do students sometimes feel they learn less from active learning? Active learning is more effortful, and that effort can feel like struggling. A 2019 Harvard study found students learned more from active sessions but felt they had learned more from lectures.

3. Does this mean memorising is bad? No. Learners need facts and vocabulary, and retrieval practice helps memory. The problem is memorisation alone, which tends to produce knowledge that is hard to apply.

4. How does STEM-Xpert put building at the centre? Through Robotics Labs, AIoT Labs, Tinkering Labs, the STEM Innovation Kit, the Crinnoboard, and Composite Skill Labs for Grade 1 to 12, with educators trained through our STEM Facilitator Course.

Fasalu Rahman CEO, STEM-Xpert

About Author Fasalu Rahman C is an education entrepreneur, STEM advocate, and the Founder & CEO of STEM-Xpert.