Engagement is not a mood you generate with enthusiasm at the front of a room. It is a measurable state in which students are thinking hard about the right things, receiving signals about how their thinking is going, and adjusting. Across every higher-education system — from a 400-seat lecture theatre in Lagos to a seminar of twelve in Manchester — the single most reliable lever for that state is well-designed formative assessment. It is the machinery that turns passive attendance into active learning.
This playbook is deliberately universal. The principles hold whether you teach on a fully online competency programme, a traditional research-intensive campus, or a resource-constrained institution where the projector fails weekly. What changes is the tooling, not the pedagogy. If you want the broader picture of how assessment fits teaching craft, browse our Teaching & Assessment collection alongside this article.
Why formative assessment beats more testing
Summative assessment measures learning; formative assessment causes it. The distinction is not about the instrument — a multiple-choice quiz can be either — but about what happens to the result. If a score is recorded and the class moves on, it is summative. If the result feeds back into the next ten minutes of teaching and the student's next attempt, it is formative. Decades of classroom research, synthesised in Black and Wiliam's foundational work on assessment for learning, converge on the same finding: frequent, low-stakes checks with actionable feedback produce larger learning gains than almost any other classroom intervention, and the gains are largest for the students who start behind.
The mechanism is straightforward. Learning requires retrieval, error, and correction. A student who never attempts recall until the final exam has no opportunity to discover what they do not know until it is too late to matter. Formative assessment manufactures those discovery moments continuously, at a stage where the cost of error is a moment of confusion rather than a failed unit. For a global teaching audience, this is liberating: you do not need expensive infrastructure to run it, only a disciplined habit of asking students to produce something you can inspect.
Low-stakes techniques that scale to large classes
The objection to formative assessment is always time and scale. How do you check the thinking of 300 students without drowning in marking? The answer is to design tasks whose feedback is fast, aggregate, or peer-driven rather than individually hand-marked. The one-minute paper — "What was the muddiest point today?" collected on paper or a form — takes students sixty seconds and gives you a diagnostic map of confusion for next week in the time it takes to skim. Think-pair-share converts a silent room into a hundred small conversations and surfaces reasoning you can sample aloud.
Concept tests, popularised in physics teaching, work at any scale: pose a conceptual multiple-choice question, have students commit to an answer, let them argue in pairs, then re-poll. The shift in the distribution between votes is itself formative evidence, and the peer instruction does most of the teaching. In classes without devices, coloured cards or a simple show of fingers work identically. The technique is agnostic to technology; what matters is that every student commits to a position and then confronts a challenge to it. For the digital equivalents, the University of New South Wales maintains a clear practitioner guide to active-learning techniques at teaching.unsw.edu.au that translates cleanly to any context. These same commitment-and-challenge moves underpin the broader engagement methods we cover in our guide to working with a teaching innovation unit.
Feedback students actually act on
Most feedback fails not because it is wrong but because it arrives too late, aims at the person rather than the work, or gives no next action. Three design rules fix this. First, feedback must be timely enough to influence the next attempt — feedback on an essay returned after the module ends is a post-mortem, not teaching. Second, it should describe the gap between current and target performance in terms of the task, not the student's character; "the claim in paragraph three is unsupported" teaches, "you're careless" does not. Third, every piece of feedback should end with a move the student can make: revise, re-attempt, compare against an exemplar.
Comparative feedback is unreasonably effective and cheap. Giving students two anonymised responses — one strong, one weak — and asking them to explain the difference against the criteria builds evaluative judgement faster than paragraphs of your commentary. It also scales, because you produce the exemplars once and reuse them. Over a semester this shifts students from asking "what mark did I get?" to "what does good look like and how far am I from it?" — the internal question that drives self-regulated learning.
Using classroom response tools without gimmickry
Polling apps, shared documents, and quiz platforms are useful precisely to the extent that they generate data you act on in the room. The failure mode is theatre: a fun poll whose results are displayed and then ignored. Before adopting any tool, ask what pedagogical decision its output will change. If the honest answer is "none", the tool is entertainment. Used well, a response system lets you run peer instruction at scale, detect a 40% error rate on a key concept in real time, and re-teach on the spot rather than discovering the gap at the exam.
Keep the technology stack lean and equitable. In many emerging-market classrooms, a proportion of students lack reliable data or devices; designing a formative routine that only works with everyone online excludes the students who most need the feedback. Paper, cards, and structured talk are not inferior fallbacks — they are robust primary methods. Reserve the digital tools for what genuinely requires them, such as instant aggregation or asynchronous retrieval practice between classes.
Aligning formative work with summative outcomes
Formative and summative assessment must rehearse the same intellectual moves. If your exam demands analysis but your weekly quizzes only test recall, students will practise the wrong skill all semester and be blindsided at the end. Constructive alignment means the verbs in your learning outcomes, your formative tasks, and your final assessment all match. Where the outcome says "evaluate", the formative work should have students evaluating repeatedly in lower-stakes settings long before the graded evaluation arrives.
This alignment is also a workload strategy. When formative tasks are miniature versions of the summative one, marking the final assessment gets faster because students arrive already competent and the distribution of quality is higher and tighter. Many academics map this deliberately using a simple grid of outcomes against tasks — the kind of curriculum-mapping discipline we explore in competency-based education models, where alignment is not optional but structural. AcademicStaff's workload calculator helps you cost these formative cycles honestly, so you can see before the semester starts whether your assessment design is sustainable across the number of students you actually teach.
Closing the loop: gathering evidence of learning
The final discipline is treating your own teaching as an object of inquiry. Formative assessment produces a stream of evidence about what students can and cannot do; capturing it turns individual lessons into a defensible account of your teaching effectiveness — invaluable for promotion, and for genuinely improving. Keep a lightweight teaching log: for each key concept, note the error rate you observed, what you changed, and whether the next cohort's errors fell. Over a few years this becomes scholarship of teaching and learning, publishable and career-advancing.
Frame this evidence around impact on students, not activity. "I introduced weekly concept tests" is an input; "the proportion of students mastering the load-bearing concept before the exam rose from 55% to 82% across two cohorts" is impact. Quality-assurance bodies worldwide — from the UK's regulators to Australia's TEQSA — increasingly expect this evidentiary stance toward teaching, and promotion committees reward it. The formative data you already collect to help students learn is the same data that documents your excellence as a teacher.
The Formative Assessment Design Grid
A one-page printable grid that maps each learning outcome to a low-stakes formative task, a feedback method, and its scale-to-class-size rating — so you can plan a full semester of engagement in under an hour.
Want the full article?
Enter your email for free access to the rest of this guide and our TEQSA resource library.
Frequently asked questions
How often should I run formative assessment in a course?
Aim for at least one low-stakes check every class or teaching session, plus a between-class retrieval task each week. Frequency matters more than elaborateness — a sixty-second muddiest-point paper every week beats one elaborate diagnostic per term, because the value is in the continuous stream of signal you and students act on.
Does formative assessment need to be graded?
No, and grading it often backfires. Attaching marks raises the stakes, discourages the productive errors that make formative work valuable, and adds marking load. If you need accountability, award small completion-based credit for attempting the task rather than grading its correctness, so students engage without fearing the cost of being wrong.
How do I do formative assessment in very large or under-resourced classes?
Use aggregate and peer-driven methods that require no individual marking: concept tests with cards or a show of hands, think-pair-share, comparative judgement of anonymised exemplars, and one-minute papers you sample rather than read exhaustively. None of these depend on technology, so they work identically in a lecture hall with 400 students or a classroom with unreliable power.
The AcademicStaff editorial team writes practical, evidence-based guidance for university staff — drawing on sector reporting, funder guidelines and the lived administrative reality of academic work. Every guide is reviewed for accuracy against current Australian higher-education practice.