How to chose the correct question type while question extraction?

VEGA AI is smart, but it works best when you give it the right type of question. If you choose the wrong question type, the AI might get confused or give wrong answers.

That’s why choosing the correct question type is one of the most important steps when you upload questions.


✅ Why It Matters

Selecting the appropriate question type helps our AI:

  • Understand the context

  • Apply the correct evaluation model

  • Generate accurate answers or feedback

  • Avoid misclassification or content rejection

If you mark it as the wrong type, it might:

  • Give the wrong answer

  • Miss the point of the question

  • Not work at all

So before uploading, make sure you pick the right type for each question.


Supported Question Types

1. Passage Question Type

  • Definition: Includes a passage body followed by one or more questions. There could be just one passage and multiple questions based on the context given in the passage.

  • Best for: Reading comprehension, evidence-based reasoning, or contextual analysis.

  • Why it matters: Our AI uses the passage to extract context and meaning. If the passage is missing or mismatched, the system may fail to extract or evaluate the question.

✅ Passage question type example:


2. MCQ (Multiple Choice Question) Type

  • Definition: A standalone question with 2 or more choices, only one of which is correct.

  • Best for: Concept-checking in math, science, grammar, or factual recall.

  • Why it matters: MCQs should be fully self-contained. If your question needs a passage, do not use MCQ — it will result in misinterpretation.

Avoid using this type for passage-based or dependent questions.

✅ MCQ question type example:


3. Numeric Question Type

  • Definition: A question where the student needs to calculate and write a number as the answer. (not multiple choice). So in simple words, the correct answer would always be a number.

  • In Numeric Question Type, there can be more than one correct answer, as long as they all have the same numerical value.

  • These are the same type of questions known as Student-Produced Responses (SPRs) in the SAT exam.

  • Best for: Math problems, physics calculations, or any computation-based tasks.

  • Why it matters: VEGA AI expects a clear numeric structure. Avoid combining this with other types like MCQ or passage-based questions.

✅ Example:


4. Open-Ended Type

  • Definition: Questions requiring a written response, typically 2–5 sentences.

  • Use this when you want the student to:

    • Explain something

    • Share their opinion

    • Write a reason or analysis

  • VEGA AI can use rubrics to evaluate open-ended or essay-type questions.

    This means the AI doesn’t just check for the right answer — it also looks at how well the response is written. It can assess:

    • Clarity of explanation

    • Structure and organization

    • Grammar and language use

    • How well the student supports their answer

    So when students write short or long answers, VEGA AI can give feedback based on real grading criteria, just like a human teacher would.

✅ Example: “Explain why El Paso became a hub for Spanish-language newspapers in the late 1800s.”


✅ Best Practices for Uploading

To ensure smooth processing and high-quality output:

Step
Guideline

Upload Quality

Ensure all files (PDFs, images) are high-resolution and easy to read.

Type Selection

Choose the correct question type for each upload. Do not mix types.

One at a Time

Upload only 1 question per extraction to maximize clarity and accuracy.

Processing Time

Most extractions take 1–2 minutes to process.

Verify Output

Double-check the extracted question text, especially LaTeX formatting, before saving.


TL;DR

Question Type
Use For

Passage

Reading comprehension

MCQ

Factual/standalone logic

Numeric

Calculation-based problems

Open-Ended

Reasoning, writing tasks

Last updated