STEAMedge attended EduCafé: Putting the STEM Education Strategic Plan into Action in Brussels on 14 April 2025, hosted by CZELO at ISELP. Speakers included representatives from the European Commission’s DG EAC, ETH Zurich, and Brno University of Technology.
Around 30% of EU students do not reach a minimum proficiency level in mathematics, and around 25% fall short in reading and science. These are not just disappointing figures. They represent a widening gap with the EU’s own 2030 target of getting underachievement below 15% (European Education Area). In 2022, only 8% of EU students reached a high level of competence in mathematics. The top performance rate has declined across the board compared to PISA 2018 (PISA 2022 Results, OECD).
The EU STEM Education Strategic Plan responds with three pillars: Lead, Level Up and Lift Barriers, with proposed targets of at least 45% of VET students in STEM, 32% in higher education, and 5% of PhD students in ICT, each with gender benchmarks attached. The ambition is real. So is the gap between ambition and implementation.
Across EU Member States, STEM does not mean the same thing. In some countries it refers narrowly to school subjects. In others it covers a broader set of competences. The addition of Arts, producing the STEAM acronym, is sometimes dismissed as a branding exercise, but there is a substantive argument behind it that deserves to be made properly.
The case is not simply that creativity is good, or that design belongs alongside engineering. It is that Arts disciplines train a specific kind of thinking: comfort with ambiguity, the ability to reframe problems, and tolerance for iterative failure. These are not soft additions to technical skill. They are the cognitive capacities that allow technical skill to remain useful as the landscape shifts. A few years ago, coding was the essential STEM skill. AI has changed that almost overnight. Curricula, by contrast, tend to lag labour market realities by five to seven years, a structural delay documented in OECD analysis of vocational education systems and acknowledged across EU education reform literature. An education system that embeds a fixed skills agenda and watches it become obsolete is not serving its students. Arts inflected thinking, comfort with uncertainty, reframing, and iteration, is one of the more durable things a curriculum can offer precisely because it is not tied to any single technology or tool.
The EU STEM Coalition is working to build shared understanding across countries, but genuine alignment remains a work in progress.
The dominant framing in policy discussions is the “leaking pipeline”: girls and women drop out of STEM at each successive stage, and the task is to plug the leaks. That framing is useful up to a point. But it locates the problem in individual women’s choices and trajectories, rather than in the environments that shape those choices. The harassment data discussed below, and the well documented culture of many STEM workplaces, suggest the pipeline does not simply leak. In places, it actively expels.
Most interventions nonetheless start too late. Research across multiple countries shows that gender stereotypes about who belongs in STEM emerge as early as age six. They are shaped by what children see around them and by the adults closest to them, and the adults closest to them are not always aware of the damage they are doing.
Parents, and mothers in particular, are among the most powerful forces shaping daughters’ early sense of what they are capable of. But the influence is more complicated than it might appear. A peer reviewed study published in Frontiers in Psychology found that direct parental support in STEM can actively backfire: when parents intervene with homework help or targeted encouragement, girls may interpret it as a signal that they are not capable enough to succeed on their own (Ertl, Luttenberger & Paechter, 2017). The support itself becomes the problem, implicitly communicating low expectations while trying to compensate for them. Well meaning parents can end up reinforcing the very gap they are trying to close.
This matters because the self concept formed in those early years is sticky. Girls who internalise the message that they are not naturally talented in mathematics, even when their grades say otherwise, carry that belief into secondary school, into university applications and into career choices. By the time formal interventions arrive, the damage is often already done.
The EU has set a target of reaching one million girls and women in STEM by 2028, with initiatives like Girls go STEM, which aims to inspire 100,000 girls aged 14 to 19, forming part of that effort. But awareness alone will not fix a pipeline that leaks at every stage, including after graduation.
The drop off begins almost immediately. Eurostat figures show that in 2021, women accounted for just 32.8% of STEM graduates across the EU. Yet in the professional world, women hold only a quarter of STEM related jobs in Europe, and the gap is even wider in ICT, where women account for just 19.4% of specialists. A degree does not guarantee entry into the field. And the further up women climb, the thinner the numbers get: women hold less than one third of higher academic positions and just 26% of decision making roles as heads of institutions (She Figures 2024, European Commission DG Research and Innovation).
When women are absent from senior levels, younger women have no one to look to, and the cycle continues. Belonging matters, and too many women do not feel it, whether in a lecture hall or a workplace.
The point the room kept returning to: it is not about what you teach. It is about how you teach it.
Most EU countries face serious shortages of qualified STEM teachers, and professional development is fragmented and unevenly distributed. The 2025 Education and Training Monitor confirms this. Evidence from Switzerland’s long term research, cited at the event, shows that more interactive teaching raises STEM engagement equally among girls and boys. European Schoolnet has long argued that teacher professional development is not a support function: it is a core driver of outcomes.
The research backs this up. The same Frontiers in Psychology study that found stereotypes corrode girls’ self concept in STEM also found that students’ interest in STEM at school, shaped significantly by how it is taught, was one of the strongest positive predictors of a young woman’s confidence in the field years later. Teachers who create conditions for genuine curiosity, rather than rote performance, are doing something that parental support and policy targets cannot replicate.
The EU STEM Education Strategic Plan includes a STEM Specialists Fellowship and tailored teacher training. These are meaningful commitments. But one off, disconnected training will not shift the dial. The evidence points clearly towards sustained, practical, peer supported professional development as the model most likely to produce durable change in classroom practice.
MyMachine is a cross level project where primary school children invent a dream machine, university students design the concept, and vocational secondary students build it. Three levels, one school year, one finished object. Each group has real stakes and a genuine reason to care about what the others produce. It is the kind of cross level collaboration the plan is trying to scale, and it was described more simply and compellingly than any policy framework could manage.
The room pushed back on the idea that AI lowers the bar for STEM skills. The consensus was the opposite. Building AI systems that are ethical, inclusive and well governed requires more people with strong STEM foundations, not fewer. The demand is shifting within STEM, towards design, interpretation and governance, which is exactly why adaptability matters so much.
The Draghi report, cited in the plan itself, identifies STEM skills as central to Europe’s strategic autonomy. That case has only strengthened as AI has accelerated. And it raises an uncomfortable corollary: an innovation agenda built on a workforce that excludes half its talent pool is not just inequitable. It is strategically incoherent.
The EU’s STEM Education Strategic Plan is serious and significant. Delivering on it requires sustained funding, genuine Member State commitment, and educators who are equipped to do the work. The gap between what the policy says and what teachers can access in terms of training, support and resources remains real and persistent. Closing that gap is not a peripheral concern. It is where implementation succeeds or fails.
At STEAMedge we equip European educators to integrate AI, sustainability, and emerging technologies into inclusive STEAM education, building the capacity to inspire students who will drive Europe’s green and digital transitions.
Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Education and Culture Executive Agency (EACEA). Neither the European Union nor the granting authority can be held responsible for them. Proposal Number: 101249439
