Build hinge questions where every distractor traps a misconception

Build hinge questions where every distractor traps a misconception

Hinge questions are the formative move that does the most work for the least class time, and the one that almost never gets planned because the bit that takes time is writing the distractors. The model is genuinely good at this. It can predict which option a Year 8 will pick if they think solids have no particles, and which one they will pick if they think melting changes the particles themselves. Worth thirty minutes of holiday week 1 for three weeks of formative assessment data in Term 3.

The prompt

You are a senior [SUBJECT] teacher in an Australian [PRIMARY / SECONDARY / TERTIARY] school. You plan hinge questions, not comprehension checks. A hinge question is the one you ask mid-lesson where the wrong answer tells you exactly which misconception is in the room and what move comes next. Today you are helping me build hinge questions for the opening lessons of Term 3.

THE CONTEXT
- Year level: [YEAR LEVEL, e.g. Year 8]
- Subject and topic: [e.g. Year 8 Science, particle model of matter]
- Curriculum frame: ACARA v9.0 [LIST THE RELEVANT CONTENT DESCRIPTORS, or paste the strand]
- State context (if it matters): [e.g. VCAA Levels 7-8 Science, NSW Stage 4, QCAA]
- The lessons you want hinge questions for: [LIST 3 TO 5 LESSONS, ONE LINE EACH, e.g. "Lesson 1: states of matter and particle arrangement"]
- The two or three biggest misconceptions you already know students hold from teaching this before: [LIST THEM, OR WRITE "I am not sure" and I will name the most common ones from the research]

PRODUCE THE FOLLOWING, IN THIS EXACT FORMAT, PER LESSON.

LESSON N: [LESSON FOCUS]

THE HINGE QUESTION
One multiple-choice question with FOUR options (A, B, C, D). It is the question that, if a student answers correctly, you can move on; if they answer wrong, you know exactly which misconception to address. Pitch it so a student with the right idea picks the correct option inside 30 seconds, with no calculation longer than two steps and no stem longer than two sentences. The question must hinge on the load-bearing idea of the lesson, not on a definition or a piece of trivia.

THE FOUR OPTIONS
A: the correct answer.
B: the answer a student picks if they hold [MISCONCEPTION 1, NAMED IN PLAIN WORDS].
C: the answer a student picks if they hold [MISCONCEPTION 2, NAMED IN PLAIN WORDS].
D: the answer a student picks if they hold [MISCONCEPTION 3, NAMED IN PLAIN WORDS, OR the "I have guessed by elimination" answer if a third distinct misconception is not honest].

Each distractor must be plausible to a student who half-understands. No throwaway options. No "none of the above". If you cannot name three distinct misconceptions for a question, drop to three options and say so.

THE TEACHER MOVE
For each wrong answer (B, C, D), write one sentence: what you actually say next, in class, the moment you see most of the room picked it. Not "we will revisit this". Specific. "Walk me through how you decided where the heat was going before it got to the water." Or: "Sketch the particles for me on the whiteboard. Where do they go when the ice melts?"

THE RETRIEVAL VARIANT
The same hinge question, rewritten so it could be asked again as a 30-second Do Now two weeks later to check whether the misconception has been corrected. Same misconception traps. Different stem and different surface context (different example, different numbers, different scenario). If a student answers correctly here but not in the original, you have evidence the correction stuck, not just the surface.

(Repeat the four blocks above for each lesson.)

A FINAL JUDGEMENT (one sentence per item, max three items)
Which of these hinge questions is the one you would not skip even if you only had time to use one per lesson, and why? If a lesson did not earn a hinge question (because it is a practice lesson, not a concept lesson), say so and recommend a worked example or an exit ticket instead.

RULES
- Australian English. ACARA v9.0 vocabulary. Year 8, not 8th grade. Term 3, not Q3.
- No em-dashes, no en-dashes anywhere. Commas and full stops only.
- Do not invent misconceptions that contradict the research. If the misconception you are naming is documented (Driver, AAAS Project 2061, the Royal Society of Chemistry misconceptions list, etc), cite the source in one line after the lesson. If you are guessing, say so.
- Do not name a distractor that is just a careless arithmetic slip. The distractors trap thinking, not typos.
- Do not use the words "leverage", "unpack", "deep dive", "robust", or "scaffold their thinking".
- Do not finish with "let me know if you want more questions". I will ask if I want more.

What it does well

Forces the model to name the misconception behind every distractor in plain words, which is the move that turns a multi-choice question into a hinge question and is the bit teachers skip when they write these alone. The retrieval variant is the quietly clever part: same trap, different surface, so a fortnight later you can tell the difference between a corrected misconception and a kid who has memorised the original question. The teacher move forces the model to commit to the next sentence you say in class, not a vague "address the misconception", and that sentence is usually closer to the right move than what you would otherwise improvise mid-lesson.

Where it falls over

The model invents misconceptions that sound plausible and are not what students actually hold. The first hinge question you write with this prompt for a topic you have taught before, sense-check the distractors against what you actually saw in the room last year. The first one for a topic you have not taught before, run it past a colleague who has. The other failure mode is forcing a hinge question into a lesson that does not have one. Skill-practice lessons, drafting lessons, reading lessons, do not always have a single conceptual fork; the prompt will produce a question anyway and it will be a comprehension check in disguise. Trust the "did not earn a hinge question" verdict in the final judgement when the model offers it.

Try it like this

For Year 8 Science, particle model of matter (the most researched misconception topic in the curriculum): paste the relevant ACARA descriptors and ask the model to cite the Driver et al research base for each named misconception. You will get three weeks of opening lessons with hinge questions you can defend to a head of faculty.

For Year 9 maths, ratio and proportion: name the additive-reasoning misconception (students using subtraction where they should use multiplication) up front, and the model will hang at least one distractor per question on it. That is the one to ask in the first week, because it is the misconception that makes the entire trigonometry unit collapse three months later.

For VCE Unit 3 Biology, photosynthesis: pre-load the model with the two big ones (plants get food from soil, light is the only input). The questions you get back are the ones you should be asking in the first week of Unit 3, not the last.

For a Year 10 English text-study lesson on theme: do not force it. The prompt will tell you the lesson did not earn a hinge question. Believe it. Use the verdict and write an exit ticket instead.

The hinge question is the cheapest formative move available, and the holiday week is when you write the ones for the term ahead.