Tech in Schools: Does Ed-Tech Actually Improve Learning?

This article may contain affiliate links. We may earn a commission if you purchase through them, at no extra cost to you.

Every generation gets its own version of the same pitch: a charismatic tech visionary stands up and tells the world that their product will revolutionise education, democratise learning, and give every child a shot at the future. Thomas Edison said it about film projectors in 1912. Steve Jobs said it about desktop computers in the 1980s. Google said it about Chromebooks. Now the same promise is being made about AI. The product changes. The playbook doesn't.

So the real question — the one that budget-conscious school administrators, parents, and policymakers actually need answered — is a blunt one: does putting technology in classrooms actually improve learning outcomes? Based on the available evidence, the honest answer is more complicated, and more sobering, than Silicon Valley would like you to believe.

This article breaks down the history of ed-tech investment, the research behind computer science education, and what schools need to understand before they hand over curriculum decisions to the next wave of tech giants promising transformation through AI.

The Ed-Tech Promise: A Century of Optimism With Mixed Results

The pattern is remarkably consistent. A powerful innovator, usually with significant financial interests at stake, identifies education as a broken system ripe for disruption. They introduce their product — film, computers, tablets, laptops, now AI — and frame it as the great equaliser. Kids will be engaged. Teachers will be empowered. Outcomes will soar.

The reality is considerably messier. A major study involving more than 500 rural schools in Peru examined what happened when every child received a laptop as part of the "One Laptop Per Child" initiative. After a decade, researchers found it did not improve learning outcomes. Children got better at using computers — which is something — but academic performance in core subjects did not meaningfully shift. Uruguay ran a similar programme and found similar results.

More pointedly, UNESCO — the United Nations' education agency — published a report finding there is little rigorous evidence that digital technology improves educational outcomes. The agency went further, noting that most of the studies that do exist on ed-tech are funded or conducted by the very companies selling tools to schools. That is a significant conflict of interest, and it deserves far more scrutiny than it typically receives in public debate.

The uncomfortable truth is that technology in classrooms tends to improve familiarity with technology. That's not nothing. But it is very different from improving reading comprehension, mathematical ability, or critical thinking — the skills that actually predict long-term outcomes for students.

Code.org and the Startup Approach to Education Reform

In the early 2010s, a former Microsoft executive named Hadi Partovi launched Code.org with his twin brother, deploying startup marketing tactics on a scale that traditional education reformers had never attempted. A viral YouTube video featuring Will.I.Am, NBA star Chris Bosh, and Mark Zuckerberg turned "learning to code" into a cultural moment. Computer science, once the domain of a narrow demographic stereotype, suddenly felt accessible and aspirational.

The signature campaign — Hour of Code — was ingeniously designed. Rather than making a slow, methodical case for curriculum reform backed by longitudinal studies, Partovi got millions of children coding simultaneously during Computer Science Education Week. The experience was gamified, joyful, and viral. Kids coded droids through mazes. They made Elsa from Frozen skate across a rink. They built animated music videos in "Dance Party" and physically danced around the classroom.

It worked as audience generation. It genuinely made coding feel approachable, particularly for students who never saw themselves in technology — girls, students from low-income households, Black and Latino students. That is a real and meaningful contribution.

But the startup instinct — move fast, scale first, study outcomes later — is also where the model runs into problems. Traditional education research operates on a simple principle: put a programme in one school, measure its outcomes rigorously, and only then expand. Code.org's approach essentially inverted that. The scale came first. The rigorous outcome data came later, and in some cases, still hasn't arrived.

Code.org also secured backing from Microsoft, Google, and Apple — companies with obvious commercial interests in normalising coding education, training future engineers, and embedding their platforms in school systems from an early age. Good intentions and business imperatives are not mutually exclusive, but they're also not the same thing.

The Academic Alternative: AP Computer Science Principles

While Code.org was generating headlines, a quieter and more methodical effort was underway in academic circles. Researcher Jane Margolis at UCLA developed a course called Exploring Computer Science, built on a fundamentally different philosophy. Rather than leading with programming syntax, the course asked broader questions: How does the internet work? What is a computer? What are the social consequences of technology?

The goal was an on-ramp — a way to bring in students who had no computers at home, no prior exposure to coding, and no reason to believe the field was for them.

Simultaneously, Jan Cooney at the National Science Foundation recognised that the diversity problem in computer science wasn't a university-level issue — it was being determined much earlier, in high schools. She worked with the College Board to develop a new AP course: AP Computer Science Principles, a deliberate contrast to the existing AP Computer Science A course, which was essentially a pure programming class with notoriously poor diversity numbers.

The new course launched with a broader mandate, was designed to complement equity-focused curricula, and has now been taken by hundreds of thousands of students. Notably, the College Board made an unusual decision: it approved packaged, pre-built curricula from multiple providers — including Code.org — allowing teachers without specialist backgrounds to deliver the course. In no other AP subject does that model exist.

The irony is that Code.org, the startup, ended up becoming one of the largest providers of AP Computer Science Principles curriculum. The two approaches — Silicon Valley viral marketing and academic equity research — converged in the classroom, even if they started from very different premises.

Ed-Tech in Schools: Pros and Cons

✅ Pros

• Accessibility: Platforms like Code.org genuinely lower the barrier to entry for students who would otherwise never encounter computer science
• Engagement: Gamified, interactive tools demonstrably increase student interest and time-on-task in the short term
• Teacher support: Packaged curricula allow non-specialist teachers to deliver subjects like computer science without years of retraining
• Scale: Digital tools can reach millions of students simultaneously at relatively low marginal cost
• Career relevance: Exposure to computing concepts provides foundational literacy for a technology-driven economy

❌ Cons

• Weak outcome evidence: UNESCO and independent studies consistently find little rigorous proof that ed-tech improves core academic outcomes
• Conflict of interest in research: Most ed-tech studies are funded by the companies selling the products — a fundamental credibility problem
• Commercial capture: When Google, Microsoft, and Apple fund curriculum development, their platforms and worldviews become embedded in public education
• Equity gaps persist: Access to hardware does not automatically translate to improved outcomes for disadvantaged students
• Skills volatility: Computer science education built around coding for software engineering looks increasingly misaligned with an AI era that is actively reducing demand for entry-level programmers
• Curriculum dependence: Packaged curricula reduce teacher autonomy and critical pedagogy in favour of corporate-designed lesson plans

Overall Rating

Ed-Tech in Schools (as currently implemented): 5.5 / 10

The concept has genuine merit. The execution is patchy, the evidence base is thin, the commercial entanglement is significant, and the track record of transformative promises is poor. There are specific bright spots — particularly in broadening access to computer science for underrepresented groups — but the sweeping claims routinely outpace the data.

What the AI Era Means for Computer Science Education

Here is where the story gets genuinely complicated for schools that spent the last decade building computer science programmes. The same tech companies that lobbied for coding education, funded curriculum development, and shaped graduation requirements are now laying off software engineers at scale and pivoting their skills requirements toward AI literacy, prompt engineering, and data interpretation.

The workforce argument for teaching kids to code — that it would produce the engineers companies desperately needed — is already weakening. Generative AI is automating significant portions of entry-level coding work. The skills gap that Code.org was built to address looks different today than it did in 2013.

That doesn't mean computer science education was a mistake. Understanding how systems work, thinking algorithmically, and grasping the social consequences of technology are durable skills. But schools that adopted computer science curricula primarily on the strength of industry workforce arguments are now watching those same industries redefine what skills they want — and preparing to run the same playbook again, this time with AI tools.

The question school districts should be asking is not "Which AI product should we adopt?" It is: Who funds the evidence behind that recommendation, and what do they stand to gain?

The Bottom Line Verdict

Technology in schools is not inherently good or bad. It is a tool, and tools need to be evaluated honestly. The century-long pattern of ed-tech promises suggests that schools and policymakers are systematically too quick to adopt, too slow to study, and too reluctant to challenge the commercial interests shaping what children learn.

For budget-conscious school administrators: demand independent outcome data before committing resources to any new platform. For parents: understand that the viral energy around coding education — and now AI education — is partly genuine enthusiasm and partly market development. For policymakers: the companies funding curriculum development are not neutral parties.

The researchers who built AP Computer Science Principles did something the startup world rarely does: they started with equity as the design principle, not virality. That approach is slower, less photogenic, and harder to fundraise around. It also produced a more durable and more honest result. That's a model worth paying attention to as the AI wave arrives in schools.

Frequently Asked Questions

Does technology in classrooms actually improve student learning outcomes?

The honest answer is: not reliably. UNESCO reviewed the global evidence and found little rigorous proof that digital technology improves educational outcomes. Studies from Peru and Uruguay found that giving every student a laptop improved their technology skills but did not improve core academic performance. Most existing ed-tech research is also funded by companies selling the products, which creates a significant bias problem.

Is computer science education in schools worth the investment?

It depends on what you're investing in and why. Broad-based computer science courses — like AP Computer Science Principles and Exploring Computer Science — that teach students how technology works, its social implications, and some foundational coding have shown value in broadening participation and building relevant literacy. Narrowly focused coding bootcamp-style programmes justified primarily by workforce arguments are more questionable, especially as AI changes what skills employers actually need.

Who is ed-tech designed for — and who does it NOT serve well?

Ed-tech tends to work best for students who already have some level of digital access, parental engagement, and self-directed learning habits. It tends to underserve students from low-income households, students with limited prior technology exposure, and students who need relational, teacher-led instruction. The equity gap between access and outcomes is one of the most consistent findings in the research.

Should schools adopt AI tools using the same model they used for coding education?

Schools should be extremely cautious about repeating the coding education playbook with AI. The same dynamics are present: large tech companies with commercial interests are funding advocacy, curriculum development, and research. The workforce arguments are being deployed again. Before adopting AI tools at scale, school districts should demand independent outcome evidence, scrutinise the funding behind any recommendations, and consider what happens to their curriculum choices if the industry pivots again in five years.