The IA evaluation is where your mark quietly changes
At some point, every IB student hits the same moment: you’ve collected the data, written the analysis, formatted the graphs… and then you open a blank page titled IA evaluation.
It feels like the emotional epilogue nobody asked for.
But examiners care about it because it answers a bigger question than “Did you get a result?” Your IA evaluation shows whether you understand how you got that result, how trustworthy it is, and what you would do differently if reality gave you one more week. That combination of honesty and control is what separates a competent IA from a convincing one.
If you want models of what “convincing” looks like, keep RevisionDojo’s coursework exemplars open in another tab as you write.
IA evaluation quick checklist (steal this)
Use this as your “did I actually evaluate?” filter:
-
I stated 2–3 strengths that genuinely improved reliability/validity
-
I identified 2–3 limitations that mattered (not random flaws)
-
For each limitation, I explained the impact on my conclusion
-
I proposed realistic improvements (specific and doable)
-
I connected the evaluation back to the research question/hypothesis
If your evaluation reads like a list that could fit any IA, it’s not done yet. For a deeper explanation of that trap, see Why most IA evaluations are too weak.
What examiners actually want from an IA evaluation
Examiners don’t want self-criticism for its own sake. They want judgment.
A high-scoring IA evaluation usually does three things in a loop:
-
Claim: name a strength/limitation.
-
Evidence: point to what you did (method choice, data pattern, source set, testing table).
-
Consequence: explain what that means for confidence in the conclusion.
That last step is the one students rush, especially under exam-season pressure. If you’re juggling coursework and finals, Can an IA be too long? The real IB answer is also a useful reminder: clarity beats volume.
Start with strengths (but make them do work)
A good IA evaluation doesn’t start with “Everything went well.” It starts with what you controlled well.
Aim for strengths that are:
-
tied to the research question
-
linked to reliability/validity (or source credibility, depending on subject)
-
supported by specifics
Examples you can adapt:
-
“Repeating trials reduced random error and made the trend more consistent across conditions.”
-
“Using two contrasting perspectives reduced single-source bias and improved balance in the argument.”
If you’re unsure what “examiner-friendly” strengths look like in practice, What IB examiners look for in a strong IA pairs well with reading an exemplar.
Then discuss limitations (the ones that change your confidence)
Here’s the rule: pick limitations that change the strength of your conclusion, not limitations that merely sound academic.
Instead of:
- “Human error.”
Try:
- “The temperature measurement uncertainty (±0.5°C) could shift calculated rate values, which weakens confidence in the exact gradient, even if the overall trend remains.”
A strong IA evaluation also avoids drama. You are not proving the investigation was “bad.” You are proving you understand uncertainty.
If you want to see how different subjects phrase this, browsing the IA guides helps you match tone and rubric language.
Improvements: make them realistic and engineered
Improvements score well when they sound like an actual next step, not a fantasy sequel.
A useful formula for your IA evaluation:
-
Limitation (specific)
-
Impact (what it did to your conclusion)
-
Improvement (exact change)
-
Why it helps (mechanism)
Example:
-
Limitation: “Timing was manual with reaction-time delay.”
-
Impact: “Adds random error to short-duration trials, weakening precision.”
-
Improvement: “Use a data logger or video frame counting.”
-
Why it helps: “Reduces human response delay and standardizes measurement.”
To sanity-check your improvement ideas, RevisionDojo’s AI Chat can be used like a strict editor: paste your limitation and ask for two improvements that are “high impact but realistic.”
Connect back to the research question (so your IA evaluation feels inevitable)
Many evaluations lose power because they float away from the research question. Bring it back.
Try sentence starters like:
-
“Taken together, these limitations suggest the conclusion is reliable for the direction of the relationship, but not the exact magnitude.”
-
“Despite these weaknesses, the evidence still supports the hypothesis because the trend persisted across repeats.”
If you want a clean example of how an evaluation sits inside a full write-up, Psychology IA IB tips: a clear guide includes an evaluation that stays anchored to the aim.
How RevisionDojo helps you write a stronger IA evaluation
A great IA evaluation is easier when you can see what strong looks like and test your own draft against criteria.
-
Use the coursework exemplars to compare tone, depth, and specificity.
-
Use the IB IA grading service to get rubric-aligned feedback fast, especially on evaluation quality.
-
Use subject rubrics and graders (for example, the IB ESS IA grader or IB Business Management IA grader) to spot what your evaluation still isn’t proving.
RevisionDojo also supports the rest of your IB workload alongside coursework: Questionbank, Study Notes, Flashcards, Predicted Papers, Mock Exams, and Tutors for when the calendar gets crowded and you need a plan that survives real life.
Closing: write the evaluation like you’re protecting your conclusion
A compelling IA evaluation isn’t about pointing out flaws to sound “critical.” It’s about protecting your conclusion by showing you understand what strengthens it, what threatens it, and what you would change next time.
If you want to level up fast, read two high-scoring examples in RevisionDojo’s coursework exemplars, then run your draft through RevisionDojo’s Grading tools and AI Chat for a rubric-focused reality check. Your IA is already an investigation. Let your evaluation prove you can think like an examiner, too.