{"course":{"slug":"ai-for-teachers","title":"AI for Teachers","url":"https://addaly.com/learn/ai-for-teachers","level":"beginner","track":"work"},"lesson":{"slug":"reading-thirty-answers-at-once","title":"Reading thirty answers at once","url":"https://addaly.com/learn/ai-for-teachers/reading-thirty-answers-at-once","position":30,"of":57,"minutes":9,"answer":"Paste a whole class's answers to one question and ask the model to group them by the error they contain (the groups are the reteaching plan).","keyIdea":"Paste a whole class's answers to one question and ask the model to group them by the error they contain (the groups are the reteaching plan), while remembering that it will find patterns in six answers as readily as in sixty."},"body":{"format":"markdown","text":"## The pile of scripts as data\n\nAfter a test you have thirty answers to question 7, and somewhere in them is the reason a third of the class got it wrong. Reading thirty answers for a pattern is slow, and by script twenty you have stopped noticing. It is the kind of reading a model is good at: it does not tire and it can hold thirty short answers in one view.\n\nThe output (three or four groups, each an error with the students' own words as examples) is a reteaching plan. This lesson is how to get it and how not to be misled by it.\n\n## Getting the answers in\n\nTyped answers from an online quiz paste straight in. Handwritten answers need transcription; the next lesson covers the phone-camera route. For a first attempt, choose a question with short answers (a line or two) and type them yourself. Thirty lines is ten minutes and it is worth doing once by hand to see what the analysis gives you.\n\nReplace names with numbers. The model does not need to know who wrote what, and the rules about student data apply.\n\n```text\nHere are 30 answers from Class 10 to: \"Explain why a ball thrown upward slows down.\" The correct idea is that gravity acts downward throughout, decelerating the ball.\nGroup the answers by the misunderstanding they show. For each group: name the misunderstanding in one sentence, list the answer numbers, and quote one answer as the clearest example. Then say which group is largest.\nDo not correct the answers. Do not invent a group that has fewer than three answers.\n```\n\n## What comes back, and what it is worth\n\nYou will get something like: nine answers say the ball \"runs out of force\"; seven say gravity \"gets stronger\" as it rises; five are right but incomplete; and so on. The first group is the classic impetus misconception and it is worth a whole starter. The second is a different error that needs a different fix. Without the grouping you would have taught one lesson to both.\n\nThat is the value, and it is real. Now the caution.\n\n## It will find patterns in anything\n\nA model asked for groups produces groups. Give it six answers and it will return three categories with percentages, and the categories will look as neat as the ones from sixty. Give it thirty answers that are all the same and it will find subtle distinctions. It has no threshold below which it says \"there is not enough here to say\". The \"no group under three answers\" line in the prompt helps; the real defence is the check question: read the answers it put in each group and see whether they belong.\n\nThat reading takes five minutes and it is where you catch the group that is two misconceptions merged, or the \"correct\" group that contains a wrong answer with the right vocabulary. Do it before you plan a lesson around the groups.\n\n## Across the whole test\n\nFor a whole paper, the spreadsheet is the tool. Marks per question per student in a grid (Google Sheets, Excel, LibreOffice Calc, all free or already on the laptop) and a row at the bottom for the class average per question. The two questions with the lowest averages are your reteaching priorities before any AI is involved. Then paste those two questions' answers for the grouping above.\n\n> Figure: Thirty answers, grouped by the error inside them. Bars (answers, out of 30): \"It runs out of force\" 9; \"Gravity gets stronger as it rises\" 7; \"The air pushes it back down\" 4; Right idea, incomplete 5; Right and complete 5. Class 10, on why a ball thrown upward slows down. The first group is the classic impetus misconception and is worth a whole starter; the second is a different error needing a different fix. Without the grouping, one lesson would have been taught to both. Read the answers inside each group before planning: the model will produce neat groups from six answers as readily as from sixty.\n\nThe model can also read the grid. \"Here is the marks grid. Which questions did students who scored well overall still get wrong?\" That finds the question that was harder than intended, or badly worded, which is a question-bank correction as much as a teaching one.\n\n## Over a term\n\nRecord the groups. After three tests you have a list of the misconceptions this class actually holds, in their words, which is the most useful planning document you own. It feeds the \"misconception\" line of every future prompt, and it is the thing the model could never have told you, because it came from your students.\n\n## The honest limit\n\nGrouping short answers works well. Grouping essays does not: the model summarises rather than analyses, and the categories become vague (\"weak structure\", \"unclear argument\") which is no better than a tired marker's impression. For long work, choose one question, one criterion, and ask about that: \"Which of these fifteen introductions state a clear position, and quote the sentence that does it?\" Narrow questions get honest answers. Broad ones get fluent ones."},"check":{"question":"A teacher pastes a class's thirty answers to one physics question and asks for the error patterns. The model returns five neat categories with percentages. Which of these should she do before planning around them?","options":[{"text":"Accept the categories, since thirty is enough for patterns and the percentages sum to 100.","correct":false,"tempting":"Neat categories and clean percentages are what the model produces regardless of whether the pattern is real; the presentation is not evidence."},{"text":"Ask the model to be more confident in its analysis and re-run it.","correct":false,"tempting":"Confidence is a tone, not a check; asking for it changes the wording and not the reliability."},{"text":"Read the answers it assigned to each category and confirm the grouping herself.","correct":true},{"text":"Re-run the analysis with the answers in a different order to see whether the categories change.","correct":false,"tempting":"Reordering is a reasonable stability check but does not verify whether the categories match what the students actually wrote."}]},"prev":{"slug":"the-student-facing-feedback-loop","title":"Feedback before submission: the student-facing loop","url":"https://addaly.com/learn/ai-for-teachers/the-student-facing-feedback-loop"},"next":{"slug":"marking-handwritten-work-with-a-phone","title":"Marking handwritten work with a phone","url":"https://addaly.com/learn/ai-for-teachers/marking-handwritten-work-with-a-phone"},"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Exam keys checked blind by a second model: 3,665 of 3,666 agreed. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'Reading thirty answers at once', in AI for Teachers. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/ai-for-teachers/reading-thirty-answers-at-once (Accessed: 13 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"Reading thirty answers at once","courseTitle":"AI for Teachers","module":{"index":4,"slug":"assessment-and-feedback","title":"Assessment, feedback and marking"},"answer":null,"keyIdea":"Paste a whole class’s answers to one question and ask the model to group them by the error they contain (the groups are the reteaching plan), while remembering that it will find patterns in six answers as readily as in sixty.","hero":{"kind":"figure","figure":{"t":"figure","label":"Figure","figure":{"type":"bars","unit":"answers, out of 30","title":"Thirty answers, grouped by the error inside them","series":[{"label":"“It runs out of force”","value":9,"highlight":true},{"label":"“Gravity gets stronger as it rises”","value":7},{"label":"“The air pushes it back down”","value":4},{"label":"Right idea, incomplete","value":5},{"label":"Right and complete","value":5}],"caption":"Class 10, on why a ball thrown upward slows down. The first group is the classic impetus misconception and is worth a whole starter; the second is a different error needing a different fix. Without the grouping, one lesson would have been taught to both. Read the answers inside each group before planning: the model will produce neat groups from six answers as readily as from sixty."},"alt":"Figure: Thirty answers, grouped by the error inside them. Bars (answers, out of 30): “It runs out of force” 9; “Gravity gets stronger as it rises” 7; “The air pushes it back down” 4; Right idea, incomplete 5; Right and complete 5. Class 10, on why a ball thrown upward slows down. The first group is the classic impetus misconception and is worth a whole starter; the second is a different error needing a different fix. Without the grouping, one lesson would have been taught to both. Read the answers inside each group before planning: the model will produce neat groups from six answers as readily as from sixty."}},"blocks":[{"t":"h2","id":"the-pile-of-scripts-as-data","runs":[{"text":"The pile of scripts as data"}]},{"t":"p","runs":[{"text":"After a test you have thirty answers to question 7, and somewhere in them is the reason a third of the class got it wrong. Reading thirty answers for a pattern is slow, and by script twenty you have stopped noticing. It is the kind of reading a model is good at: it does not tire and it can hold thirty short answers in one view."}]},{"t":"p","runs":[{"text":"The output (three or four groups, each an error with the students’ own words as examples) is a reteaching plan. This lesson is how to get it and how not to be misled by it."}]},{"t":"h2","id":"getting-the-answers-in","runs":[{"text":"Getting the answers in"}]},{"t":"p","runs":[{"text":"Typed answers from an online quiz paste straight in. Handwritten answers need transcription; the next lesson covers the phone-camera route. For a first attempt, choose a question with short answers (a line or two) and type them yourself. Thirty lines is ten minutes and it is worth doing once by hand to see what the analysis gives you."}]},{"t":"p","runs":[{"text":"Replace names with numbers. The model does not need to know who wrote what, and the rules about student data apply."}]},{"t":"code","lang":"text","label":"Text","text":"Here are 30 answers from Class 10 to: \"Explain why a ball thrown upward slows down.\" The correct idea is that gravity acts downward throughout, decelerating the ball.\nGroup the answers by the misunderstanding they show. For each group: name the misunderstanding in one sentence, list the answer numbers, and quote one answer as the clearest example. Then say which group is largest.\nDo not correct the answers. Do not invent a group that has fewer than three answers."},{"t":"h2","id":"what-comes-back-and-what-it-is-worth","runs":[{"text":"What comes back, and what it is worth"}]},{"t":"p","runs":[{"text":"You will get something like: nine answers say the ball “runs out of force”; seven say gravity “gets stronger” as it rises; five are right but incomplete; and so on. The first group is the classic impetus misconception and it is worth a whole starter. The second is a different error that needs a different fix. Without the grouping you would have taught one lesson to both."}]},{"t":"p","runs":[{"text":"That is the value, and it is real. Now the caution."}]},{"t":"h2","id":"it-will-find-patterns-in-anything","runs":[{"text":"It will find patterns in anything"}]},{"t":"p","runs":[{"text":"A model asked for groups produces groups. Give it six answers and it will return three categories with percentages, and the categories will look as neat as the ones from sixty. Give it thirty answers that are all the same and it will find subtle distinctions. It has no threshold below which it says “there is not enough here to say”. The “no group under three answers” line in the prompt helps; the real defence is the check question: read the answers it put in each group and see whether they belong."}]},{"t":"p","runs":[{"text":"That reading takes five minutes and it is where you catch the group that is two misconceptions merged, or the “correct” group that contains a wrong answer with the right vocabulary. Do it before you plan a lesson around the groups."}]},{"t":"h2","id":"across-the-whole-test","runs":[{"text":"Across the whole test"}]},{"t":"p","runs":[{"text":"For a whole paper, the spreadsheet is the tool. Marks per question per student in a grid (Google Sheets, Excel, LibreOffice Calc, all free or already on the laptop) and a row at the bottom for the class average per question. The two questions with the lowest averages are your reteaching priorities before any AI is involved. Then paste those two questions’ answers for the grouping above."}]},{"t":"p","runs":[{"text":"The model can also read the grid. “Here is the marks grid. Which questions did students who scored well overall still get wrong?” That finds the question that was harder than intended, or badly worded, which is a question-bank correction as much as a teaching one."}]},{"t":"h2","id":"over-a-term","runs":[{"text":"Over a term"}]},{"t":"p","runs":[{"text":"Record the groups. After three tests you have a list of the misconceptions this class actually holds, in their words, which is the most useful planning document you own. It feeds the “misconception” line of every future prompt, and it is the thing the model could never have told you, because it came from your students."}]},{"t":"h2","id":"the-honest-limit","runs":[{"text":"The honest limit"}],"role":"limit"},{"t":"p","runs":[{"text":"Grouping short answers works well. Grouping essays does not: the model summarises rather than analyses, and the categories become vague (“weak structure”, “unclear argument”) which is no better than a tired marker’s impression. For long work, choose one question, one criterion, and ask about that: “Which of these fifteen introductions state a clear position, and quote the sentence that does it?” Narrow questions get honest answers. Broad ones get fluent ones."}]}],"check":{"question":[{"text":"A teacher pastes a class’s thirty answers to one physics question and asks for the error patterns. The model returns five neat categories with percentages. Which of these should she do before planning around them?"}],"options":[{"text":[{"text":"Accept the categories, since thirty is enough for patterns and the percentages sum to 100."}],"correct":false,"tempting":[{"text":"Neat categories and clean percentages are what the model produces regardless of whether the pattern is real; the presentation is not evidence."}]},{"text":[{"text":"Ask the model to be more confident in its analysis and re-run it."}],"correct":false,"tempting":[{"text":"Confidence is a tone, not a check; asking for it changes the wording and not the reliability."}]},{"text":[{"text":"Read the answers it assigned to each category and confirm the grouping herself."}],"correct":true,"tempting":null},{"text":[{"text":"Re-run the analysis with the answers in a different order to see whether the categories change."}],"correct":false,"tempting":[{"text":"Reordering is a reasonable stability check but does not verify whether the categories match what the students actually wrote."}]}],"order":[3,0,1,2],"why":null},"prev":{"slug":"the-student-facing-feedback-loop","title":"Feedback before submission: the student-facing loop","minutes":8},"next":{"slug":"marking-handwritten-work-with-a-phone","title":"Marking handwritten work with a phone","minutes":8}}}