How much data should I include in my IA analysis?

Enough to support your claims, and not one figure more. A reliable rule is: one major piece of evidence per major claim, plus whatever is needed to show reliability (like variability or uncertainty). If a table or graph isn’t referenced and interpreted in the text, it probably doesn’t belong in the main body of the IA. When in doubt, move raw or repetitive material out of the spotlight and keep the narrative clean. This is also why reading How many tables should an IA include? can help you cut without panic. Your examiner should feel guided, not buried.

Should raw data go in the main body of an IA?

Usually, no. The main body of an IA is for processed data that you actively use to argue something: averages, trends, comparisons, model outputs, and visuals that reveal a pattern. Raw data can be valuable for transparency, but it often overwhelms the reader and interrupts the reasoning. A better approach is to summarize the raw data into one clean table or graph and then interpret it. If your subject requires raw data access, keep it organized and clearly labeled so it supports credibility without stealing attention. The core principle is simple: the main body is where your thinking lives.

How do I write analysis that earns marks instead of sounding obvious?

Move from “what happened” to “why it happened” and “why we should trust it.” Start by naming the pattern precisely (rate of change, correlation strength, plateau, turning point), not just saying it increased or decreased. Then connect the pattern to theory or context you’ve already introduced, showing you understand mechanisms rather than just outcomes. Finally, evaluate what the limitations do to your confidence: does uncertainty weaken the conclusion or merely soften it? This triad (pattern, explanation, evaluation) is the heartbeat of strong IA writing. If you want to train this skill, studying annotated models in RevisionDojo’s coursework exemplars is one of the fastest ways.

IA Data Analysis: Use Evidence Like an Examiner

RevisionDojo Team

•9/15/2025•8 min read

This site uses cookie tracking technologies. Learn more in our Cookie Policy.

IA Data Analysis: Use Evidence Like an Examiner

RevisionDojo Team

•9/15/2025•8 min read

The night before an IA deadline, data starts to look like a foreign language. Numbers blur. Graphs multiply. And suddenly you’re tempted to paste in one more chart because it feels like work.

But examiners don’t grade effort. They grade thinking.

In a strong IA, data isn’t decoration. It’s the backbone of your argument. Whether you’re handling trial results in a science IA, source patterns in a History IA, or a model fit in a Math IA, the goal is the same: use data to make your conclusion feel inevitable.

If you want models of what that looks like in real student work, start with RevisionDojo’s coursework exemplars. Seeing how high-scoring students talk to their data changes everything.

A messy spreadsheet vs one clear graph comic

IA data checklist (the fast version)

Before you polish wording, run this checklist through your IA analysis:

Your data is relevant to the research question or aim of the IA
You present processed data in the main body (raw data goes elsewhere)
Every figure has labels, units, and a clear caption
You interpret patterns and meaning, not just what the graph “shows”
You connect each claim back to your IA focus
You acknowledge limitations, anomalies, and uncertainty with impact

For a deeper write-up framework, keep 10 best practices to write up data and results in your IB IA open while you edit.

Why data handling makes or breaks an IA

A quiet truth about the IA is that most students can collect “fine” data. The mark difference comes from what happens next.

Examiners are watching for four signals:

Clarity: can they understand your evidence in seconds?
Accuracy: are calculations, labels, and visuals consistent?
Analysis: do you explain why the trend exists?
Evaluation: do you discuss what makes you more or less confident?

If you’re not sure what “analysis” vs “description” looks like, How to analyze graphs and results in the IB Math IA is a useful transfer skill even outside math.

Collect reliable data (so your IA has something to stand on)

Your IA analysis can’t outsmart shaky evidence. Reliability looks different by subject, but the principle is stable: make choices that reduce doubt.

In a science IA, this often means enough trials, careful control of variables, and consistent measurement.
In a humanities IA, it means balanced sourcing and clearly justified selection of evidence.
In a math IA, it means datasets large enough to show patterns and not just coincidences.

Whatever your subject, document where the data came from and why it’s appropriate. If you’re still shaping your focus, How to write a strong IA research question helps you build a question that your data can actually answer.

Present data like you respect the reader

A clean visual is a form of generosity. It says: “I’ve done the sorting, so you can do the thinking.”

In a high-scoring IA, visuals are purposeful:

Use titles that state what the reader is about to learn
Label axes, include units, and keep scaling sensible
Choose graph types based on purpose (trend, comparison, distribution)
Avoid duplicating the same information in multiple forms

If you want a dedicated guide on visual choices, read How to present data visually in your IA. And when you’re deciding how much to include, How many graphs should an IA include? saves you from the “more must be better” trap.

Analyze, don’t narrate: the sentence upgrade your IA needs

A common IA weakness is writing captions in paragraph form.

Narration: “Figure 2 shows that results increase as X increases.”
Analysis: “As X increases, Y rises at a diminishing rate, suggesting a limiting factor. This supports the expectation from theory because __, but the flattening after __ indicates __.”

A practical editing move: after every result, force yourself to write three lines:

So what? (meaning)
Why might that be? (mechanism/theory)
What does it do to my confidence? (uncertainty/limitations)

That simple rhythm turns data into argument, which is what an examiner is hunting for.

Link every data point back to your IA focus

Your IA is not a scrapbook. Every table, quote, or graph has to earn its space.

After you present evidence, add one explicit line that connects it to your aim:

“This trend matters because it suggests __, which directly addresses my research question by __.”

When students skip this, examiners do extra work to infer relevance. And if an examiner has to work to understand the point, you usually lose marks.

If your draft is ballooning, Can an IA be too long? The real IB answer is a calm reminder that clarity beats volume.

Evaluate limitations and anomalies like an adult

In real research, clean data is rare. A strong IA doesn’t hide that. It explains it.

Instead of listing generic issues (“human error,” “small sample size”), talk about:

What specifically went wrong or could be biased
How it likely affected the trend or conclusion
What realistic improvement would reduce the impact

If your evaluation feels thin, Why most IA evaluations are too weak is worth a read, then follow with How to write a compelling IA evaluation section.

RevisionDojo tools that make IA data work easier

At some point, every IA becomes an editing problem: too many visuals, not enough interpretation, unclear links to criteria. This is where RevisionDojo quietly saves hours.

The Coursework Exemplars library gives you real, high-scoring IA models so you can benchmark your structure and depth.
RevisionDojo’s Grading tools help you check whether your analysis and evaluation actually match rubric language.
AI Chat can help you test your reasoning: “What alternative explanation could fit this trend?” (You still write the final logic.)
Study Notes and Flashcards support the theory behind your analysis, so you’re not hand-waving.
For exam season, Questionbank, Predicted Papers, and Mock Exams keep coursework and revision aligned, so your IA doesn’t steal all your time.
If you’re stuck, Tutors can help you refine methodology and interpret results without crossing authenticity lines.

Conclusion: Make your IA data tell a story with a point

A high-scoring IA isn’t the one with the most data. It’s the one where the data clearly answers the question, and the analysis shows mature judgment about what the evidence can and cannot prove.

If you want your IA to read like an argument an examiner can trust, build on clear presentation, disciplined interpretation, and specific evaluation. Then sanity-check your structure against real examples in RevisionDojo’s coursework exemplars and tighten your draft with rubric-focused feedback using RevisionDojo’s Grading tools and AI Chat.

Your data already has a story. Your job is to make it readable, honest, and impossible to ignore.

This site uses cookie tracking technologies. Learn more in our Cookie Policy.

IA Data Analysis: Use Evidence Like an Examiner | RevisionDojo

The night before an IA deadline, data starts to look like a foreign language. Numbers blur. Graphs multiply. And suddenly you’re tempted to paste in one more chart because it feels like work.

But examiners don’t grade effort. They grade thinking.

If you want models of what that looks like in real student work, start with RevisionDojo’s coursework exemplars. Seeing how high-scoring students talk to their data changes everything.