IB Extended Essay Research Question: 4-Step Guide

Quick answer: A 2027 EE research question must be written as a question, clearly stated and focused, grounded in the selected DP subject or subjects, open to amendment, and capable of supporting critical analysis, evaluation, synthesis and a line of argument within 4,000 words.

Confirm which EE model applies

The revised model is first assessed in May 2027 and uses subject-focused and interdisciplinary pathways. Students assessed in 2026 must follow the earlier guide confirmed by their coordinator. The question cannot be judged separately from the applicable pathway and subject-specific guidance.

Topic, title and research question are not synonyms

• Topic: the issue or context being investigated.
• Research question: the question the research and essay will answer.
• Title: a short descriptive statement on the finished essay; it is not normally phrased as the research question.

“Corporate culture” is an interest area, not a research question. Adding a company name creates a case, but the student still needs a question, analytical lens, evidence plan and purpose.

The four-step research-question workflow

Step 1: Confirm a focused topic and academic lens

First complete the EE topic selection and feasibility check. Then state:

1. the selected pathway and subject or subjects;
2. the defined object, case or issue;
3. the subject knowledge or methods that will be used;
4. why the question is worth investigating.

For an interdisciplinary EE, identify what integrating the two subjects makes possible. Two parallel mini-essays do not create an interdisciplinary line of argument.

Step 2: Run a preliminary investigation

Inspect real sources before finalising the wording. Every 2027 EE requires secondary research, and the subject guidance determines whether student-collected primary research is appropriate.

Use a simple evidence map:

Evidence check	Question to answer
Availability	Can the student access enough relevant material?
Quality	Are the sources credible for this discipline?
Range	Can more than one perspective, explanation or finding be tested?
Method	Can the evidence be analysed using appropriate subject methods?
Limits	What uncertainty, bias or access problem will affect the answer?

If the source scan changes the likely argument, change the working question. That is a sign of research, not failure.

Step 3: Draft a real question

The 2027 guide says the research question should invite critical examination and encourage analysis, evaluation and synthesis rather than description, unsupported generalisation or value judgement. It should also avoid a double-barrelled question and a self-evident answer.

Question stems such as “How significant…?”, “How successful…?” and “To what extent…?” can help, but no stem is a scoring formula. The method and evidence must be able to answer what the wording asks.

These hypothetical transformations illustrate the process rather than ready-to-submit questions:

Weak starting point	More focused working direction	What still needs validation
How does advertising affect people?	How significant was one defined campaign in changing one measurable response among a named audience?	Business or another subject lens, evidence and causal limits
What does a novel say about identity?	How does a defined authorial choice construct one aspect of identity in a specified work?	Language A text eligibility and literary method
Does temperature affect a reaction?	How does a feasible temperature range affect a defined reaction measure under controlled conditions?	Science safety, method, uncertainty and subject guidance

Do not copy an example merely because it sounds academic. A question only works when the student can defend its subject fit, sources, methods and boundaries.

Step 4: Apply six quality tests

1. Question test: Is it actually written as one question rather than a topic, hypothesis or intention?
2. Subject test: Can the applicable subject concepts and methods answer it?
3. Evidence test: Have credible, accessible sources been inspected rather than assumed?
4. Argument test: Is there room for a reasoned, non-obvious answer supported by analysis?
5. Scope test: Can it be handled comfortably within 4,000 words without becoming superficial?
6. Integrity test: Is it distinct from the student's IA and other assessed work, and is the decision authentically the student's?

If the scope test fails, use the EE topic narrowing diagnostic.

What not to force into every question

There is no universal requirement for:

• a “To what extent” opening;
• one independent and one dependent variable;
• a country and date range;
• primary data collected by the student;
• a comparison between exactly two cases;
• a question that never changes.

Those elements are useful only when the subject, evidence and method require them.

Keep a question-change log

For each substantial version, record the date, wording, evidence that triggered the change, scope removed or added and supervisor discussion. This supports deliberate research and later reflection without turning reflection into a chronological activity log.

Final supervisor conversation

Bring the question with a one-paragraph rationale, five-source sample, proposed method and current limitations. Ask whether the pathway, subject fit, feasibility, ethics and non-duplication are clear—not whether the question is guaranteed to score well.

Continue with the complete IB Extended Essay workflow, or review AI use for EE and IA feedback before using any external tool.

StudyIB is independent and is not affiliated with or endorsed by the International Baccalaureate Organization. Confirm the applicable guide and school requirements.