Guest blog by Abel Capellan Almonte, IPP ’25
I entered IPP with a simple story in my head: students “don’t like” STEM. After months of PDIA, I think that story is convenient but wrong. The problem is not a single decision made by a teenager. It is a system that quietly pushes them toward safer, more visible paths. If we want higher STEM participation, we must argue with the system, not with the student.
PDIA begins by insisting we construct the problem before we “solve” it. At first this felt slow. But the fishbone diagram changed the conversation. Once we mapped causes information gaps, weak hands-on experiences, math anxiety, parental concerns, unclear labor-market signals the narrative shifted from “students lack motivation” to “our system hides the path.”
This is not semantics. When we blame the student, the remedy is motivational speech. When we see structural causes, the remedy is many small moves across the pathway: better signaling, micro-experiments in classrooms, parent communication, and credible role models. PDIA convinced me that precision in the problem is an ethical duty: it protects learners from being blamed for our design failures.
I ran small, low-cost probes: short STEM Ambassador talks from young professionals, micro-labs with simple kits or data from local life, guidance micro-bursts inside homeroom, and parent touchpoints via WhatsApp and one-page briefs. None of these were heroic. But together, they produced better questions. Students moved from “Is STEM too hard?” to “Which math should I take next term?” That shift from fear to planning is a leading indicator of future behavior.
Critics may say: “Talks and micro-labs are not policy.” I disagree. In complex systems, small, repeated signals can rewire expectations. An ambassador who looks like the students lowers the social distance to STEM. A 40-minute lab replaces abstraction with touch. A 15-minute myth-busting session prevents a bad rumor from setting like concrete. And a short parent brief respects the fact that home is an authorizing space. PDIA taught me to value these moves not as decoration, but as evidence-generating steps. Each probe answered a learning question and prepared a stronger next step.
Another hard lesson: success did not depend on one “change champion.” It depended on sequencing authorizers. A school director allowed the timetable slot; department heads allowed the micro-labs; ministry peers signaled alignment with curricula; parents allowed the after-school time. PDIA’s idea of “earning room for the next inch” is more realistic than the usual reform romance. You don’t need unanimous approval for a pilot you need enough authorizing space for a small move that produces visible learning
I could import a beautiful STEM toolkit from abroad, but students listen when the example is theirs. Our best micro-lab used local data (traffic, dengue, electricity). Suddenly ,STEM was not a foreign brand; it was a tool to read our own city. In my view, context is not a constraint, it is the engine of relevance. PDIA’s insistence on crawling the design space ,testing multiple small options instead of betting on one grand solution, helped me find the versions that survived contact with reality.
I used to plan for approval. Now I design for learning. Each action must answer a specific question:
- Will a near-peer ambassador increase student question quality?
- Do shorter guidance bursts reach more students than on a single career day?
- Which parent concerns can be reduced with a one-page brief?
This mindset reduces performance theater. If an action does not teach us something useful, it is not a PDIA step; it is a show. PDIA turned my calendar from “events” to hypotheses.
I will not claim big numbers. But the direction is strong: better attendance in optional sessions, teachers volunteering to co-run micro-labs, parents asking for concrete checklists, and students discussing prerequisite choices for the next term. These are micro-signals of pathway formation. In a pipeline problem, such signals matter more than one-off excitement.
How this approach will guide my next move
- Institutionalize a “STEM Activation Kit.” A repeatable package of any school can run in two weeks:
- one ambassador talk
- one micro-lab
- two guidance micro-bursts
- one parent brief
- Clarify the three tracks. Dominican students face technical, technical-superior, and university routes. I will publish simple, visual guides showing prerequisites, scholarship windows, and job signals for each so STEM looks like a set of doors, not a wall
- Lightweight data loops. Short pulse surveys after activities and a quarterly review with school leaders to adapt frequency, timing, and content.
- Sustain a mentor pool. A small list of volunteers (universities and industry) who can give 30–40 minutes per month. This is affordable and places living proof inside the school.
Some readers may still think these are small tactics for a big national issue. PDIA gives a different standard: seriousness is measured by learning speed and adaptive integrity, not by the size of the press release. If an initiative cannot adjust after new evidence, it is not serious; it is fragile. By contrast, small steps that re-target based on feedback are robust and scale-ready. We don’t scale programs; we scale solutions that survive.
Words for fellow PDIA practitioners
- Don’t romanticize your first idea. Marry the learning loop, not the prototype.
- Make the problem visible. A fishbone diagram is a social technology; it builds alignment faster than a memo.
- Ask for the next inch. Shop for small permissions that unlock evidence, then use the evidence to negotiate the next inch.
- Write short learning notes. Memory is a public good; notes protect momentum when people rotate.
- Stay local. Local data and faces move hearts and calendars
If I have to summarize my learning in one idea, it is this: low STEM participation is not a single choice, it is a system of messages. Students hear those messages at home (“play it safe”), in classrooms (“the lab can wait”), from peers (“STEM is for a few”), and even from our institutions
(complex pathways, confused scholarships, unclear jobs). PDIA gave me a method to argue with those messages, one small, evidence-generating step at a time.
This approach changes the unit of progress. Instead of counting events, I now look for shifts in expectations: students asking about prerequisites; teachers requesting time for micro-labs; parents forwarding the one-page brief; a director offering a regular slot in the timetable. These
are small signals, yes but they are leading indicators of a pipeline that can grow. When such signals multiply and stabilize, policy has something to scale.
The broader argument is about equity. STEM should not be a narrow door for the already confident. In the Dominican Republic, many talented students live close to the margin of information, time, and money. A low-cost activation loop ambassador talk + micro-lab + guidance bursts + parent brief makes the path visible and doable without waiting for perfect resources. This is not a replacement for bigger reforms (curricula, labs, teacher development), but it buys time and builds demand for them. In PDIA terms, it expands the authorizing space from the ground up.
Sustainability requires institutions, not heroes. My next move is to package the steps into a simple kit any school can run in two weeks, with short instructions, a calendar template, and a feedback form. The point is not a perfect manual; the point is a reliable learning loop that schools own, adapt, and improve. If each cycle answers one question and documents the result, the system becomes smarter every quarter without big budgets.
There are risks. Momentum can fade, staff can rotate, and exams can squeeze the timetable. That is why I will keep light data (pulse surveys, attendance, quality of student questions) and quarterly reviews with school leaders. If one element stalls, we adjust sequence or format. PDIA
is not a promise that every step will work; it is a promise that we will learn fast enough to avoid wasting a year.
Finally, PDIA changed me. I used to design for approval; now I design for learning. I used to defend my plan; now I defend my hypothesis and evidence. I am not a native English speaker and sometimes my words are simple, but the practice is clear: listen carefully, try small, learn honestly, adapt quickly, and respect people’s constraints. If we keep that discipline, STEM becomes a normal option not a special program and more Dominican students will see themselves in it, choose it, and stay. That is a future worth working for.
This is a blog series written by the alumni of the Implementing Public Policy Executive Education Program at the Harvard Kennedy School. 36 Participants successfully completed this 5-month hybrid program in September 2025. These are their learning journey stories.