Find the Gap: How to Identify Misunderstandings from Student Answers
'Find the Gap' is a tool that analyzes the answers provided by learners and helps teachers identify misunderstandings, then suggests targeted practice. You insert textual answers into the 'Source material' field, formulate a goal or desired result, and launch generation. As a result, you receive one text report in a separate panel, which can be copied or downloaded in TXT format. In this article we will explore how to prepare input data, how to read the result, and what limitations are important to consider so that the tool brings real benefit and does not create an illusion of automatic diagnostics.
Open tool: Knowledge gap finder ↗What the Tool Does and When It Is Useful
'Find the Gap' accepts two text fields: source material (student answers) and goal or desired result. Based on this data it generates a report listing likely misunderstandings and suggesting topics for review or exercises. The tool does not assign grades, does not compare students against each other, and does not replace the teacher -- it is an auxiliary means for quick diagnostics.
Typical scenario: you have collected written answers from a group on the topic 'Percentages' and want to understand which errors repeat most often. You paste the answers into the 'Source material' field, state a goal -- for example, 'learn to correctly calculate the final price after consecutive discounts' -- and launch generation. After a short while you receive a structured list of problem areas.
The tool works only with text. It does not analyze audio, video or images, does not check spelling and does not correct grammar. If student answers were oral, they must be transcribed manually first. There is also no capability to upload a document -- text is entered directly into the field.
Source material: 'I think percentages are added directly: a 20% discount, then another 10% -- that is just 30%'. Goal: 'Learn to correctly calculate the final price after consecutive discounts'.
How to Prepare Source Data so the Result Is Useful
The quality of the output depends directly on the quality of the input. In the 'Source material' field it is better to insert connected student answers, keeping their formulations. If you shorten or paraphrase the answers, the tool may not see the real error. For example, the phrases 'percentages are added' and 'percentages are added on' may be perceived differently, so keep the original words.
Text volume is limited: 8000 characters for the first field and 1000 characters for the details field. If you have more answers, split them into several runs or select the most illustrative ones. Do not try to fit an entire test into one request -- it is better to focus on 5-10 answers that vividly demonstrate typical errors.
In the 'Goal or desired result' field formulate specifically: not 'improve knowledge', but 'learn to convert percentages to fractions and back'. The more precise the goal, the more relevant the suggested exercises will be. If the goal is vague, the tool may give general recommendations that will not help in your situation.
Instead of 'students do not know percentages' write: 'Goal -- learn to calculate the final cost of a product after two consecutive discounts, without adding percentages directly'.
Typical Errors in Source Data and How to Avoid Them
One of the common problems is incomplete answers. If a student wrote only '30%', the tool will not understand the context. Always add the question or task so the system sees what exactly was being checked. For example: 'Question: What is the final discount when 20% and 10% are applied? Student answer: 30%.'
Another mistake -- mixing answers from different students without labels. If you insert phrases from five people in a row, the tool may decide this is one contradictory answer. Separate answers with a blank line or mark 'Student 1:', 'Student 2:'. This is not a mandatory requirement, but significantly improves the readability of the output.
Also avoid including your own comments in the source data field. If you add 'here is an error', the tool may simply repeat your assessment rather than finding the gap on its own. Leave only raw student answers and a clear goal.
Bad: '20%+10%=30% -- incorrect'. Good: 'Student: 20%+10%=30%. Question: What is the final discount?'
How to Read and Check the Result
After generation you will receive a text report. Usually it contains a list of presumed misunderstandings and exercise recommendations. Do not treat this list as absolute truth. The tool can make mistakes, especially if the answers were ambiguous or contained typos. Your task is to critically evaluate each item.
Check that the indicated error is actually present in the answers. For example, if the tool writes 'student does not understand that percentages cannot be added', go back to the original text and make sure there is exactly such a statement. If the answers were ambiguous, the tool may have made an incorrect conclusion.
Compare the suggested exercises with your curriculum. Some recommendations may be too general ('review the topic of percentages') or, conversely, too narrow. Use them as a starting point for your own lesson planning, not as a ready-made lesson plan.
The tool may suggest: 'Exercise on distinguishing additive and multiplicative percentages'. Check if you have materials for such an exercise, and adapt them to the group's level.
Tool Limitations and What It Cannot Do
'Find the Gap' is not an automatic knowledge assessment system. It does not assign scores, does not compare students against each other and does not track progress over time. This is a one-time text analysis. If you need dynamics, you will have to save reports manually and compare them yourself.
The tool does not work with images, audio or video. It cannot transcribe spoken speech, does not analyze charts and does not check pronunciation. All input data must be in the form of plain text. There is also no capability to upload a document -- only to paste text into the field.
Finally, the tool does not guarantee that the identified gaps are actually the cause of errors. It merely indicates likely misunderstandings based on language patterns. The final diagnosis is made by the teacher, using experience and knowledge of the students.
If a student erred due to carelessness, not misunderstanding, the tool may still indicate a gap in the topic. This is normal -- use the result as a hypothesis.
Example Use Case: Analyzing a Percentage Error
Suppose you have collected student answers to the question: 'An item costs 1000 rubles. First a discount of 20%, then another 10%. What is the price of the item now?' One student answered: '700 rubles, because 20+10=30%, and 1000-300=700'. Another wrote: '720 rubles, I calculated consecutively'. You paste both answers into the 'Source material' field and state a goal: 'Learn to calculate consecutive discounts without adding percentages directly'.
The tool may output a report: 'Identified misunderstanding: student adds percentages directly. Recommended exercise: distinguishing additive and multiplicative percentages. Also repeat that the second discount is applied to the already reduced price'. This is a specific and useful hint.
You check: indeed, the first student erred exactly that way. The second answer is correct, but the tool may not mention it, as it is looking for gaps. Based on the report you prepare a mini-lesson: you show on a numeric example why 20% of 1000 and 10% of 800 are different quantities. This is the targeted practice.
Source material: 'Student A: 700 rub, 20+10=30%. Student B: 720 rub, 1000*0.8=800, 800*0.9=720'. Goal: 'Learn consecutive discounts'. Result: the tool identified the additive error and suggested an exercise.
How to Fix an Unsuccessful Generation Result
If the tool's output turned out to be useless or incorrect, first check the source data. Most often the problem is that the text was too short, contained typos or lacked a clear question. Add context: what was asked, what the student answered. Make sure the goal is formulated specifically.
Try splitting a large text into parts and run analysis separately for each error. For example, first analyze only percentage answers, then -- fractions. This will give more precise recommendations. Do not be afraid to experiment with the goal formulation: sometimes just reordering words helps the tool better understand the task.
Remember that the model can make mistakes. If the result contradicts your experience, do not use it blindly. It is better to rephrase the query or analyze answers manually. The tool is an assistant, not a replacement for pedagogical judgment.
If the tool gave general phrases, try adding more specificity to the goal: 'Learn to convert 20% to a coefficient 0.8 and apply it consecutively'.
The 'Find the Gap' tool can become a useful assistant in diagnosing typical errors if used consciously. Prepare quality source data, formulate specific goals and critically check the results. Remember the limitations: it is a text analyzer that does not replace the teacher and does not give guarantees. Use it as a source of hypotheses for planning targeted practice. In the percentages example we saw how an additive error can be identified and an exercise proposed. Such an approach saves time and helps focus on real student gaps.
Open tool: Knowledge gap finder ↗Interface translations are machine-assisted and may need refinement.