A hinge question prompt that diagnoses student misconceptions

A hinge question prompt that diagnoses student misconceptions

The classroom-ready move with AI this term is not to generate more quiz questions, but to use it to find the misconceptions hiding inside the ones you already ask.

The prompt

You are an expert in [SUBJECT] pedagogy with deep knowledge of how Australian students in [YEAR LEVEL] typically misunderstand the idea below.

Concept I am teaching: [CONCEPT IN ONE PLAIN SENTENCE]
Curriculum link: [ACARA v9.0 content descriptor, VCE study design dot point, or your own learning intention]
When I will use it: 5 minutes mid-lesson, every student answers together (fingers, mini-whiteboards, or Plickers).

Write ONE hinge question:
1. Multiple choice, with exactly four options labelled A, B, C, D.
2. The correct answer is unambiguous and defensible to a parent.
3. Each of the three incorrect options must map to a specific, named misconception that students in this year level genuinely hold. Do not write random distractors. Name the misconception in plain English in italics directly after the option.
4. Answerable in under 60 seconds, no calculator, no device.
5. Australian English, Australian contexts where natural.

Then, in a separate "Teacher use" block, tell me:
a. What I should do next if more than 30 percent of the class picks each wrong option.
b. The follow-up question I should ask if the class splits roughly evenly between two options.
c. One thing this question will NOT tell me about student understanding.

What it does well

Forces the model to commit to a named misconception for each distractor, which is the actual diagnostic value of a hinge question and the thing most teacher-made multiple-choice items skip. The "Teacher use" block turns it from a quiz item into an actionable mid-lesson decision tree.

Where it falls over

The model does not know your students. The misconceptions it lists are "common in the literature", not "common in 8C". You still have to read the names of the kids who picked C and ask yourself whether the model called it right, and you should bin any distractor that does not match a misconception you have actually seen in the room.

Try it like this

For Year 7 maths (equivalent fractions): expect distractors mapped to "added the same number to numerator and denominator", "bigger numerator means bigger fraction", and "treated the fraction bar as a subtraction".

For VCE English (analytical paragraphs): expect distractors mapped to "summary masquerading as analysis", "personal opinion without textual evidence", and "two arguments crammed into one topic sentence".

If you only have five minutes today, run it once for the lesson you are teaching tomorrow morning. One good hinge question beats a worksheet.

This is the kind of practice the AITSL Teacher Standards review (Sunday's Daily Brief) is likely to formalise under "safe, effective and ethical use".