You can feel it when an IA is almost there.
The graphs look clean. The argument mostly holds. The formatting is finally behaving. And yet, the best IB students still spend serious time on one section that feels less glamorous than calculations or analysis: evaluation.
In a top-level IA, evaluation is where you stop performing competence and start showing judgment. Examiners are not only asking, “What did you do?” They are asking, “How well did it work, and how do you know?” That is why evaluation often becomes the quiet separator between a mid-range IA and one that earns the highest bands.
If you want to see what this looks like in real, high-scoring coursework, start with RevisionDojo’s Coursework Exemplars. Reading one excellent evaluation paragraph can save you hours of guessing.
IA evaluation quick checklist (save this before you edit)
A strong IA evaluation is short, specific, and connected to marks. Use this as your final pass:
-
Name 2-4 meaningful limitations (method, data, sources, assumptions)
-
Explain impact on reliability/validity/interpretation (not just “it’s a limitation”)
-
Propose realistic improvements that match each limitation
-
Acknowledge alternative explanations or interpretations
-
Show what you learned about the process (reflection, not excuses)
For a bigger picture of how the whole IA is assessed and moderated, it also helps to read Are IAs Externally Moderated? What IB Students Must Know. It clarifies why marker-friendly evaluation matters.
Why evaluation matters in a top-level IA
Think of evaluation as the moment you prove you understand what your work can and cannot claim.
In most subjects, the IA rewards students who can:
-
separate strong evidence from weak evidence
-
explain uncertainty instead of hiding it
-
defend decisions using criteria language (reliability, validity, bias, assumptions, representativeness)
-
suggest improvements that would actually change the quality of results
In other words, evaluation is not “being negative.” It is being precise.
If you want a broad roadmap for the entire process (topic choice to final write-up), keep IB IA Guides: Internal Assessment Structure, Rubrics & Tools open while you work. It keeps your evaluation anchored to what the IB actually assesses.
The anatomy of a high-scoring IA evaluation
A top evaluation paragraph usually follows a repeatable pattern:
Limitation (specific) → Impact (why it matters) → Improvement (how to fix + why it helps) → Resulting effect (what would likely change)
That last part is where many students level up: you do not just propose a change, you explain the consequence of that change.
To compare how strong students structure this across subjects, browse subject sets like Maths AA IA exemplars or IB Economics IA Grade 7 Examples. Different subjects, same evaluation mindset.
Write limitations like a scientist, not a confession
Many students write limitations as if they are trying to reduce blame. But the best IA evaluation reads more like a lab report or a careful historian: calm, evidence-based, and connected to outcomes.
What “specific” actually looks like
-
Too weak: “Human error affected results.”
-
Stronger: “Manual timing introduced reaction-time uncertainty (approximately ±0.2 s), which likely widened the spread of results and reduced confidence in the trend.”
-
Too weak: “Sources may be biased.”
-
Stronger: “Two of the three sources were produced by stakeholders with direct incentives, so claims about impact may be overstated; this reduces reliability when comparing perspectives.”
A great shortcut is to ask RevisionDojo’s AI Chat a targeted question like: “Is this limitation measurable, and have I explained its impact?” Then tighten the sentence until it answers yes.
Explain impact: turn “what went wrong” into “what it changes”
Evaluation earns marks when you explain the effect on your conclusion.
A useful test for every evaluation sentence in your IA:
-
Does it change how confident I am in the result?
-
Does it change how generalizable my conclusion is?
-
Does it change whether my method truly measured what I claimed?
If you are unsure whether you are writing impact or just listing problems, read High-Scoring IA Reflections: Examples and Lessons. It models how reflection stays analytical.
Improvements: match the fix to the limitation
The easiest way to lose evaluation marks is to offer improvements that are generic.
“Use better equipment” sounds sensible, but it is not evaluative unless you specify what equipment, what it changes, and why it improves the IA.
A stronger improvement sentence includes two extra details
-
Mechanism: how the change reduces error/bias/uncertainty
-
Outcome: what that would do to reliability/validity/interpretation
Example:
-
Weak: “Repeat the experiment more.”
-
Strong: “Increase trials from 3 to 8 and randomize order to reduce the influence of outliers and fatigue effects, improving reliability of the mean and tightening uncertainty.”
For method-specific expectations (especially in sciences), it can help to cross-check with IA Methodology: Write a Clear, Replicable Section. Often, your best evaluation points are already hiding inside your methodology choices.
Weave evaluation through the IA (not only at the end)
A high-scoring IA usually signals evaluation in three places:
-
During method/design: justify choices and flag likely weaknesses early
-
During analysis: mention uncertainty, assumptions, anomalies, and alternative interpretations
-
In the conclusion: summarize the most important limitations and best improvements
If your draft is getting long, you can still do this without adding words. Often you replace narration with judgment.
If word count and scope are becoming a problem, Can an IA Be Too Long? The Real IB Answer is a helpful reset. It explains why “more writing” is not the same as “more marks.”
Using RevisionDojo to upgrade evaluation fast
When time is tight, you need feedback loops.
Here is a practical workflow many students use inside RevisionDojo:
-
Use Study Notes to repair any missing theory that your evaluation depends on (so you can explain impact correctly).
-
Drill evaluation-style thinking with Questionbank prompts that force you to “evaluate” and “discuss,” not just calculate.
-
Check your draft with Grading tools (rubric-aligned) to see whether your evaluation is specific enough to earn top descriptors.
-
Pull structure ideas from the Coursework Library and Coursework Exemplars.
-
Use Flashcards for key terms like validity, reliability, bias, uncertainty, and assumptions so your language stays precise.
-
When exam season hits, keep the loop tight with Mock Exams and Predicted Papers so your coursework effort supports exam performance too.
If you want a deeper understanding of how moderation rewards clarity, How IA Moderation Works (and How to Protect Your Grade) connects evaluation directly to what moderators can justify with evidence.
Conclusion: the IA is judged by how you think, not just what you did
A top-level IA is not the one with zero flaws. It is the one where the student can explain flaws with precision, show how those flaws affect conclusions, and propose improvements that genuinely strengthen the investigation or argument.
If you want your evaluation to sound like a high scorer, do one thing today: open RevisionDojo’s Coursework Exemplars, read one excellent IA evaluation section, then rewrite your own using the same limitation → impact → improvement logic. Pair that with RevisionDojo’s AI Chat and Grading tools for fast iteration, and you will feel your IA move from “finished” to “credible.”