Everything in this article follows from one sentence in the IB's own guidance:
The focus of the assessment of the IA is on students' ability to justify their methodological decision-making, not on their ability to design the perfect study.
Read it again with your own draft in mind. You are not being marked on whether your study is clever. You are being marked on whether you said why.
The test to run on every paragraph
Highlight every because, since, in order to and so that.
If a paragraph has none, it is describing. That is fine where the criterion says describe, and it is costing you marks everywhere else.
Most middle-band methodology sections are accurate, clear, complete and reasonless. The fix is not more content. It is one clause per decision.
Compare these:
The sample will consist of 20 participants aged 14 to 16 from one secondary school, recruited through opportunity sampling.
The sample will consist of 20 participants aged 14 to 16 from one secondary school, recruited through opportunity sampling because the population of interest is defined by recent arrival at that specific school, which makes a full sampling frame impossible to construct.
The second is fourteen words longer and a band higher. Nothing got more sophisticated. A reason appeared.
Justifying the method
The criterion wants one thing: why is this method the most appropriate way to investigate this particular question?
Not why the method is good in general. Why it fits your aim and your population.
The logic runs in this direction. Your aim requires a particular kind of evidence. Only one of the four methods produces that kind of evidence well. Therefore this method.
Because the aim is to examine how newly arrived students experience belonging, the data required is detailed verbal accounts of their own perceptions, which a semi-structured interview produces and a questionnaire cannot.
Add a second sentence on why it suits the population, and that bullet is done.
Justifying the sample
Three decisions, three reasons.
The technique. Name it specifically, then say why it was preferred over an alternative. Opportunity sampling is not a weakness to apologise for. It is a choice with reasons, and saying what those reasons are earns more than pretending you would have run a stratified sample.
The characteristics. Who qualifies, and why those criteria. If participants must have joined within the past twelve months, say what that criterion protects.
The number. How many, and why that many. A reason as simple as needing enough transcripts to identify patterns while keeping the interviews analysable is a reason, and it counts.
One distinction worth keeping clean, because it is a common slip. Sampling is how people enter your study. Allocation, if you are running an experiment, is how they enter a condition. Those are separate decisions doing separate jobs, and each needs its own reason.
Justifying the procedure
Procedure is broader than most people assume. It covers sampling technique, sample characteristics, design if the study is experimental, setting, and the process itself.
The mark is not for listing steps. It is for explaining them.
So write each step and attach its reason. Why this setting. Why this order. Why this duration. Why this instruction given at this moment.
A procedure section that reads as a clear, complete set of instructions with no reasons attached sits at 3 to 4. That is the single most common shape of a competent methodology section, and it is fixable in twenty minutes.
Ethics: the one bullet with a different rule
There is something unusual about the ethics bullet, and it is worth knowing before you write.
It is the only top-band bullet in the entire rubric that stops at described. It does not climb to explained.
What lifts it instead is one word: explicitly.
The considerations that matter must be described and tied, in so many words, to your own investigation. An implied link is not enough.
| Band | What it looks like | You are here if |
|---|---|---|
| 1 to 2 | Considerations described but not connected to your investigation at all | Your ethics paragraph would fit any IA ever written |
| 3 to 4 | Relevant considerations described, but some are not linked to your study | One or two are tied to your study and the rest are generic filler |
| 5 to 6 | Relevant considerations described and explicitly linked, with steps | Every ethical point names something about your own study, and each has an action attached |
The practical instruction: name your participants, name your topic, say what would actually be done.
Not "informed consent would be obtained." Instead, something that could only have been written about your study: participants would be told that the interview concerns their experience of settling into the school, that they may decline any question about friendships, and that recordings would be deleted after transcription.
Two tests. Could this sentence appear in someone else's IA unchanged? If yes, rewrite it. And does every ethical point have an action attached, rather than a principle stated?
Ethics is methodology, not decoration
This is the move that separates a strong ethics paragraph from a compliant one, and it is available to you if you have room.
Ethical decisions have methodological consequences. Anonymity can make sensitive self-report more honest. The right to withdraw can produce attrition that changes your sample. Full disclosure of the aim protects autonomy and invites demand characteristics.
Naming one of those connections shows that you understand ethics as part of research design rather than a box to tick.
The same logic runs through your tool
Criterion C is the same skill applied to a different object.
You created a tool. Every choice inside it was a decision: these items, this scale, this wording, this order, this length. Each one needs a reason connected to your aim.
And one specific sentence people leave out. Claim your tool would be effective, and say why. The criterion asks for a tool that fits your aim and would actually work, and the difference between the middle and the top band here is often a single passage saying so.
The four fastest fixes
If you are short on time before submission, these are ranked by how much each strengthens a descriptor for the time it takes.
- Add a reason to every step of your procedure.
- Name your own study inside every ethical point.
- Claim your tool is effective, and say why.
- Write the sentence in your discussion saying how a second method would increase understanding.
None of these requires new research or a better idea. All four are sentences you either wrote or did not.
The self-check
Print your methodology and data collection sections. Highlight every because, since, in order to and so that.
If a paragraph has no highlighting, you have found the paragraph costing you marks.
If you want the full sentence banks, band descriptors for all twelve bullets and a complete model proposal, that is what the PsychExcel Complete IA Handbook is built for.