High-Impact Teaching
Strategies Toolkit

Here's a subtlety that changes how you read the research: nearly every teaching practice has some positive effect, so 'it works!' isn't a high bar. The real question is relative — which practices work most, for the time and effort they take? That's what effect sizes help with, and what this toolkit prioritizes. The aim isn't to do everything; it's to spend your limited time and energy on the highest-impact moves.

Hattie himself says there's 'no fixed recipe' and 'no set of principles apply to all learning for all students' — and researchers have rightly critiqued the precision of effect-size rankings. So don't treat any number as gospel or chase the top of a list. The deeper, more durable lesson of this research is a mindset: make learning visible, gather evidence of what your teaching is actually doing, and adjust. Effect sizes point you toward promising practices; checking their impact in your room is what makes you great.

Teacher clarity — being explicit about the learning goal and what success looks like — is one of the most reliably effective and least complicated moves in teaching. When students understand what they're trying to learn and can picture a good outcome (through clear success criteria and exemplars), they focus their effort, monitor their own progress, and learn more. Don't let the objective live only in your head or on a slide they ignore — make it genuinely clear and keep returning to it.

Explicit instruction is often misunderstood as dull, teacher-talking-at-students lecturing — but done well, it's highly interactive, full of questioning, modelling, and practice. And the evidence is clear that for novice learners tackling new material, fully guided explicit instruction outperforms minimally-guided discovery or inquiry, which tends to overload and confuse. As students gain expertise, you gradually release support (§07). Model it fully first — 'I do' — then hand over the reins step by step.

Rosenshine's principles come from three converging sources — cognitive science, studies of effective teachers, and research on scaffolding — which is why they're so robust. Tom Sherrington groups them into four strands: sequencing & modelling, questioning, reviewing, and stages of practice. The crucial caveat (Sherrington's): don't turn them into a lesson-observation checklist to tick off. They describe the underlying architecture of effective instruction across subjects and contexts — a way of thinking, not a script to perform.

Cognitive scientist Daniel Willingham's line captures the heart of it: students remember what they think hard about. Because working memory is limited, dumping too much on students at once — too many steps, too much extraneous detail, too little modelling — overloads them and stops learning cold. The fixes are practical: break new material into small steps, remove clutter, use worked examples so novices aren't flailing, and gradually withdraw support as they master each piece. Teach in a way the brain can actually process.

The single most transferable move from learning science into the classroom is regular, low-stakes retrieval: a few recall questions to start class, a quick 'write everything you remember,' a mini-quiz on last week's content. Retrieving information (rather than re-presenting it) strengthens memory far more than review, and spacing it out beats cramming it in. This is the same science in our Study Skills toolkit — and The Learning Scientists have teacher-ready resources on it.

Two small shifts have an outsized effect. First, wait time: most teachers wait less than a second after asking a question, so only the fastest hands go up — pausing three to five seconds lets every student think and dramatically improves the answers you get. Second, cold call ('hands up only to ask questions, not answer them'): calling on students at random signals that everyone is expected to think and lets you see what the whole class understands, not just the eager few. Done warmly, they transform participation.

A striking research finding: when students get a grade and a comment, most read only the grade and ignore the comment — so comment-only feedback often improves learning more than a grade with feedback attached. The wider principle (Wiliam): feedback only works if it changes what the student does, which means it must be specific, actionable, and require them to do something with it. Great feedback should be 'more work for the recipient than the donor' — and it means giving less, better feedback, not more.

The core of formative assessment is simple but transformative: don't teach a whole lesson (or unit) and only discover at the end who understood — gather evidence as you go (through questioning, mini whiteboards, exit tickets) and adjust in real time. As Dylan Wiliam puts it, formative assessment is a process of eliciting evidence of learning and acting on it immediately. It also means activating students to assess their own and each other's work — turning them into partners in the learning, not just recipients of it.

Rosenshine found effective teachers aim for a high success rate (around 80%) during practice — high enough that students are genuinely succeeding and building confidence, but challenging enough to require effort. Practicing while making lots of errors just rehearses the errors. So guide practice until students can succeed, then release them to independent work, and practice basics to fluency — because automatic basics free up mental space for the harder thinking you actually want.

It's easy to design lessons that are busy, fun, and engaging on the surface — and where it's genuinely unclear what students are learning. Wiliam warns that activity-based teaching too often optimizes for engagement rather than learning. The better test isn't 'are students busy and having fun?' but 'is every student thinking hard about the thing I want them to learn?' Design for that — cognitive engagement, not just behavioral or emotional engagement — and use techniques (§09) that get all students thinking, not just the visible volunteers.

Metacognition and self-regulated learning — helping students plan, monitor, and evaluate their own learning — carry a strong evidence base (the EEF rates it among the highest-impact, lowest-cost approaches). And unlike content, these skills transfer: a student who learns to check their own understanding, choose good strategies, and reflect carries that into every future subject and challenge. Model your own thinking aloud, teach strategies explicitly rather than hoping students pick them up, and see our Study Skills & Executive Function toolkit.

The evidence puts positive teacher-student relationships among the strongest influences on achievement — and the magic combination is warmth plus high expectations, not a choice between being caring and being demanding. Students rise to the level of belief their teacher holds for them (and, unfortunately, sink to low expectations too). Know your students, build genuine relationships, create a climate where mistakes are safe, and hold every one of them to high expectations you truly believe they can meet. The strategies land better on that foundation.

The shift from 'differentiation' (often misread as endlessly individualized, watered-down tasks) to adaptive teaching is an important one: the goal is to hold all students to high, grade-level expectations while flexibly varying the support that gets them there — scaffolds, models, pre-teaching, targeted help, and responsive reteaching. It's more sustainable and more equitable than creating separate, lesser work for some students. For specific populations, our Special Education, Multilingual Learners, and Gifted toolkits go deeper.

Despite being believed by a large majority of teachers, the idea that matching instruction to students' 'learning styles' (visual, auditory, kinesthetic) improves learning has no credible evidence — study after study finds no benefit. It's a well-meaning neuromyth. This doesn't mean ignore how you present material (using words and visuals — dual coding — genuinely helps everyone); it means don't sort students into style boxes or feel guilty for not doing so. Dropping debunked ideas frees time and attention for what actually works.

Two hopeful truths close out the strategies. First, teaching expertise isn't fixed — it's built through deliberate practice: working on specific moves, getting feedback, and refining over time (which is exactly what this toolkit's strategies give you to practice). Second, it grows fastest collectively: the most powerful force in Hattie's entire dataset is teachers who, together, believe they can succeed with every student and organize their work around that belief. Don't teach in isolation — get better, together.

The research on improving teaching is consistent: the biggest wins come from building teachers' collective belief and skill over time — through sustained, practice-based professional learning and instructional coaching — not from one-off workshops, mandates, or turning research frameworks (like Rosenshine's) into observation checklists to enforce. Give teachers evidence-informed practices, time to practice them, feedback from a skilled coach, and a collaborative culture — and teaching gets better. See our Instructional Coaching and School Culture resources as they publish.

For the strongest evidence base, the Education Endowment Foundation's Teaching & Learning Toolkit summarizes what works and at what cost; Visible Learning holds Hattie's rankings; and Dylan Wiliam and The Learning Scientists turn research into practice. But the most important resource is your own classroom: try high-impact strategies, check what they do for your students, and refine. Start at K12academics.com for the wider world of education.

Editable versions of these guides are available on request — see §26, Stay Connected.