All Psychology guides

Psychology IA Criteria Guide

Plan a research proposal that addresses a real problem.

For assessment in 2027. Explain the problem, proposed method, data collection tool and possible findings. You are proposing research rather than reporting the results of a completed experiment.

Assessment criteria

What each criterion measures, how many marks it carries, and what stronger evidence usually looks like.

Criterion names, marks and mark bands follow the IB subject guide. The band descriptions are our own summaries, so check your subject guide for the exact wording.

Criterion A: Introduction (6 marks)

Connect a focused proposal to a real problem, population and relevant research.

Stronger evidence
The aim is precise. The effect on the chosen population is explained, and two relevant studies are explained and connected to the proposal.
Early-stage evidence
The aim is unclear or broad, the problem is only named, and the research support is weak or incomplete.

Criterion B: Research methodology (6 marks)

Explain the proposed method, procedure and ethical choices.

Stronger evidence
The method and procedure are explained, and relevant ethical choices are tied directly to the proposal.
Early-stage evidence
The account of the method has mistakes, the procedure has gaps, and ethical points are disconnected from the proposal.

Criterion C: Data collection (6 marks)

Create a suitable tool to measure behaviour and explain design choices and collection challenges.

Stronger evidence
The tool would measure behaviour effectively. Its design choices and likely collection problems are explained in relation to the question.
Early-stage evidence
A suitable tool is supplied, but it has mistakes. Design choices and possible collection problems receive little relevant detail.

Criterion D: Discussion (6 marks)

Consider possible findings, researcher bias and an additional method.

Stronger evidence
Possible findings and practical implications are explained. Researcher bias is examined, and an additional method is justified by what it could help us understand.
Early-stage evidence
Possible findings receive little explanation. Bias is only named, and the additional method is weakly connected.

Mark bands

MarksWhat the draft shows
0Nothing in the work reaches the lowest band.
1-2The aim is unclear or broad, the problem is only named, and the research support is weak or incomplete.
3-4The aim has some focus and the problem is described. Research is relevant, but its explanation or connection to the proposal is incomplete.
5-6The aim is precise. The effect on the chosen population is explained, and two relevant studies are explained and connected to the proposal.

MarksWhat the draft shows
0Nothing in the work reaches the lowest band.
1-2The account of the method has mistakes, the procedure has gaps, and ethical points are disconnected from the proposal.
3-4The chosen method and ethical points are relevant, but the procedure or links to this proposal need more explanation.
5-6The method and procedure are explained, and relevant ethical choices are tied directly to the proposal.

MarksWhat the draft shows
0Nothing in the work reaches the lowest band.
1-2A suitable tool is supplied, but it has mistakes. Design choices and possible collection problems receive little relevant detail.
3-4The tool suits the proposal. Its design choices and likely collection problems are described in relation to the question.
5-6The tool would measure behaviour effectively. Its design choices and likely collection problems are explained in relation to the question.

MarksWhat the draft shows
0Nothing in the work reaches the lowest band.
1-2Possible findings receive little explanation. Bias is only named, and the additional method is weakly connected.
3-4Possible findings and relevant bias are described. An additional method is discussed, but its value or practical implications need stronger links.
5-6Possible findings and practical implications are explained. Researcher bias is examined, and an additional method is justified by what it could help us understand.

Want to know where your IA sits in these bands? Upload your draft and get a mark for each criterion.

Grade my IA

How to build it

Most weak drafts fail on order, not on ideas.

  1. 1

    Define the problem and aim

    Explain the effect on your chosen population. Connect a focused aim or question to the findings and conclusions of two relevant studies.

  2. 2

    Explain the proposed method

    Describe the procedure clearly and explain why the method fits. Tie ethical choices to this particular proposal.

  3. 3

    Create the collection tool

    Include the questions, observation schedule or other tool you would use. Explain the design choices and likely difficulties in collecting data.

  4. 4

    Discuss possible findings

    Explain what potential findings could mean for policy or practice. Discuss researcher bias and how an additional method could improve understanding.

Before you submit

  • The problem, population and research question fit together.
  • Two relevant studies support the proposal.
  • The actual data collection tool is included and its design choices are explained.
  • The discussion covers possible findings, practical implications, bias and another method.

Quick wins

  • Check every method choice against your aim.
  • Explain how each ethical decision protects the people involved.
  • Keep possible findings separate from results that have actually been collected.

Check your Psychology research proposal

Get marks and comments against the 2027 IB Psychology IA criteria. The result includes a mark for every criterion, evidence from your draft, and the changes to make first.

  1. 1

    Upload your IA draft as a PDF

  2. 2

    Get a mark and feedback for every IA criterion

  3. 3

    Revise and grade the new version

A mark for every criterion
A mark for every criterionSee where the draft is now, criterion by criterion.
Comments on your sentences
Comments on your sentencesEach comment points at the passage it is about.
What to fix first
What to fix firstConcrete next edits instead of "improve your analysis".