LearnGrokExit tickets are useful only if they change what you teach next. These prompts help you turn a pile of short answers into misconception groups, a pupil follow-up list and a focused reteaching sequence. Use them after teaching a topic, when the next lesson is still adjustable.
Do not paste full pupil names where a pupil code will do. Use the same codes in every prompt so that the groupings, tasks and feedback record match.
Key point
Start with the raw answers
A correct or incorrect answer is not a misconception. Group pupils only when their written method or explanation gives evidence of the same misunderstanding.
Prepare the exit-ticket document
Before using the first prompt, make one working document named Exit-ticket responses: [topic], [class], [date]. Include:
- the learning intention and success criteria
- the exact exit-ticket question
- the expected answer or worked method
- each response beside a pupil code
- blanks, absent pupils and illegible responses
Keep the wording pupils actually used. For mathematics, include calculations and diagrams, not just their final number. For a written subject, include the sentence or phrase that supports your judgement. A transcription that removes the method usually removes the evidence you need.
If a response could be an accidental slip, record it as such. A pupil who writes the wrong number after using a sound method needs a different response from a pupil whose method shows they have selected the wrong operation.
Watch out
Do not turn silence into a diagnosis
A blank answer may mean absence, lack of time, uncertainty, a reading problem or missing equipment. Collect one more piece of evidence before allocating a conceptual misconception.
Use the prompts in order
- Run Group responses by misconception first. It gives you a class-level view and protects against the common mistake of grouping every wrong answer together.
- Paste that output into Build a pupil follow-up list. This turns patterns into actions for named pupil codes. It should identify a small teacher group, pupils who need a quick individual check, and pupils who can continue independently.
- Choose the highest-priority group and use Plan the reteaching sequence. The goal is not to repeat yesterday's explanation. Plan a different example, representation or contrast that addresses the misunderstanding shown.
- Use Make tasks for different evidence groups to prepare the actual questions. The teacher group needs practice that exposes its reasoning. Pupils who were secure need work that still checks the same learning intention.
- Run Check the plan against the evidence before teaching. This is particularly useful when the output has grouped many pupils, or when you are tempted to create several interventions from one short question.
You do not have to use every prompt. If the first analysis shows one clear misconception across the class, move directly to the reteaching plan. If responses are mostly blank or ambiguous, use the follow-up list and evidence-gathering task before planning a full intervention.
Check whether the analysis is wrong
Read the output beside the raw exit tickets. Check every quoted pupil code and response. The model can make a neat category from weak evidence, especially when answers are brief.
Use this table when reviewing the result.
| If you see this | Treat it as | What to do next |
|---|---|---|
| Same final wrong answer, different working | Possibly different misconceptions | Split the group by method, not answer |
| Correct answer with no working | Limited evidence | Ask a hinge or follow-up question |
| Blank, crossed-out or unfinished work | Missing evidence | Collect a short response next lesson |
| A group with several possible explanations | An uncertain pattern | Write a question that distinguishes the explanations |
Check
The group has earned its label
You should be able to point to at least two responses that show the same faulty idea, not merely the same result.
Check the hinge question as carefully as the analysis. Answer each option yourself from the information pupils have been taught. If two options could be defended because the wording is vague, rewrite the question. A hinge question is only useful if a response leads to a clear next action.
Also check proportionality. Two pupils with uncertain evidence may need a one-minute conversation, not a separate lesson task. A widespread misconception that blocks the next objective may justify a short whole-class reset before groups begin.
Keep a short record after the lesson
Use the fast feedback record while pupils complete the planned task. Mark secure, not yet secure, or unclear from what you see, rather than from the original exit ticket. This gives you a cleaner starting point for the following lesson.
Note
Keep the source with the output
Save the raw exit-ticket document with the analysis. You need the original wording when you revisit a grouping or explain why a pupil was selected for follow-up.
For current information about using the service and features that may vary by account or version, check the xAI documentation overview.
When the prompts do not work
If the analysis is vague, do not ask for a more confident answer. Add the missing materials: the exact question, answer key, worked methods and pupil wording. If the output creates too many groups, ask for a maximum of three instructional groups and require it to place uncertain cases separately. If it misses an important pattern, paste two or three representative responses and ask what evidence would confirm or challenge your interpretation.
When the exit ticket cannot distinguish the cause of an error, use the next lesson to gather better evidence. A short, well-chosen question is more useful than a detailed but unsupported label.
