Gerald Ajam

From educational principle to usable resource

What a principle has to become before a teacher can use it.

Status From practice

Published November 2026

All practice cases

Institutional detail is generalised. Teams, officers and schools are not named, and internal targets are left out.

Context

From 2016 to 2019 I developed geography resources for a national curriculum team. Much of that work was for a national learning platform, piloted in 2017 and then opened to every school. Our job was to have curriculum-aligned lessons ready for students and teachers when it did.

In 2017 I wrote our strategy for those lessons. In 2019 I led the development of print and digital materials for a new lower secondary syllabus. Throughout, the work ran on a small set of principles from the learning sciences and multimedia research.

Those principles later went into a guide shared by resource developers across subjects. In 2026 I rebuilt its sixty principles as a public reference on this site. This case runs from the first lesson template to that reference.

The question

What does a research principle have to become before a teacher, or the person writing for one, can use it?

What made it difficult

The deadline was fixed and the scope was wide. Lessons had to be on the platform when it opened, from lower secondary to pre-university geography.

The platform set the form. It could mark closed questions automatically and nothing else. Any feedback a student got without a teacher had to hang on a question with a fixed set of answers.

We expected students to use the lessons in the gaps of a day, on a bus or between other things. So each lesson had to fit into about twenty minutes.

Many teachers were wary of a platform that had been billed as a “game-changer”. Our own guidance to colleagues said to expect scepticism and to acknowledge it.

And the principles didn’t say what to do. Each came from studies with particular learners under particular conditions. None told a writer what to put in the second question of a lesson on weather and climate.

What we noticed

Teachers asked questions no principle answers. How much homework will this add? How do we know a student really finished? Why don’t the questions look like exam questions?

Their critiques were specific. A short lesson can’t build understanding on its own. There were too many words, especially in the extension material. Most questions tested knowledge, and skills were thin. Some wanted badges, because otherwise, they said, students wouldn’t do the work.

Writing a resource up against the principles showed how unevenly they applied. Several had nothing to act on, because the lessons had no narration. One exposed a weakness: nothing tested the extension material, so students would probably skim it.

Our own rules pulled against each other. To help students finish, we asked every part of a lesson to make low demands. Elsewhere we built practice to be spaced, interleaved and varied, which is effortful on purpose.

The principles on our checklist turned up almost one for one in the shared guide a few years later, beside new ones on visual, interface and inclusive design. Each carried the same authority as the next.

When I checked their sources in 2026, some weren’t findings at all. A well-known diagram of how readers scan a page had no traceable source. Advice on another reading pattern pointed the opposite way from the research group whose study gave it its name. Fonts sold for dyslexic readers don’t help them.

Judgement

Give each principle a place in the lesson. Every content lesson had the same parts, each with one job: an inquiry question as the title, a short start linking to what students already knew, the media, three questions with feedback, a wrap-up and somewhere to go further. A writer who filled in the parts was applying the principles without having to recall them.

Make the feedback do the teaching. Since only closed questions could be marked, we built the options around likely misconceptions and wrote the feedback to model a geographer’s thinking, in the language students would need in an essay. A wrong answer led to an explanation of how to think about the question.

Answer the teacher’s question first. Our guidance said the lessons should enhance or replace existing work and never add to it. Time spent learning outside class should stay the same or fall.

Write the reasoning down beside the resource. We documented resources against the principles they drew on, on a checklist with a column for how each had been handled. For the new syllabus, every line of digital lessons came with design notes, so a later team could see why each choice was made.

Grade the canon. When I rebuilt the shared principles in 2026, I marked every one on its face as replicated, supported, convention or contested, with one line saying why. A reader deserves to know what a claim is worth at the moment they meet it.

What changed

A strategy paper in 2017 set out two kinds of lesson, drawing on Laurillard’s types of learning: lessons for taking in content and lessons for practising skills. It fixed the structure every content lesson would follow.

We built a series of map skills lessons for spaced, interleaved and varied practice across map types, with feedback written in the language of a geography essay.

We tested early versions with students. In 2018 I showed other resource developers how to prototype, test and trial materials before full production.

We wrote guidance for colleagues on how to talk to teachers about the lessons, with answers to the questions teachers had actually asked.

For the new lower secondary syllabus, we piloted a coursebook with seven schools, tested each line of digital lessons with users before full production and drafted a teachers’ guide to using the materials.

In 2026 I rebuilt the sixty shared principles as a public guide, each with its evidence mark. Where the evidence contradicts the advice as usually taught, the guide says so and puts what is known in its place.

What happened

By April 2018, as the platform reached schools, our lessons were on it for every lower and upper secondary geography course. The coursebook pilot in 2019 drew useful suggestions from students, and our review judged that it supported the design choices.

The graded guide is public. Of its sixty principles, twenty-six are replicated findings, twenty-three rest on sources narrower than the claims made of them, seven are craft conventions and four are contested.

What I can’t tell you is whether students learned more geography because of any of this. We tracked what we produced and tested whether it was usable. We never compared learning with and without the lessons, and I don’t know how teachers adapted them once they were in every school.

What transferred

  1. A principle reaches a teacher as a structure, a default or a piece of feedback.

    It almost never arrives as a principle.

  2. Write the why beside the what.

    The next team inherits the artefact. Without the reasoning, they inherit it as a rule.

  3. The platform’s limits shape the form as much as the evidence does.

    Name the limit, then design inside it on purpose.

  4. Grade what you hand over.

    A list that sets a finding and a rumour in the same type will be used as if both were true.

Singapore