Data analysis is the moment your Science Extended Essay stops being a pile of numbers and becomes a claim you can defend.
Most IB students don’t struggle because they “can’t do maths.” They struggle because they can’t tell which numbers matter, which method is justified, and how to explain the difference between a real pattern and a lucky coincidence. In IB Sciences, that gap is often the difference between “I did an experiment” and “I made an argument.”
This guide breaks your EE data analysis into seven practical steps you can repeat. It’s built for the way IB Sciences marks are awarded: clarity, appropriate processing, critical thinking, and a line of reasoning that never loses the research question.
A quick checklist before you start analyzing
Use this as your pre-analysis sanity check for IB Sciences EEs:
-
Your research question is specific enough to test
-
You can name the independent, dependent, and controlled variables
-
You have raw data with units (and uncertainties where relevant)
-
You know what “processed data” will look like (means, rates, gradients, indices)
-
You can explain why your statistical test or model fits your data
-
You have a plan to evaluate reliability, validity, and limitations
If you’re still refining your structure, RevisionDojo’s IB Extended Essay Guides and the breakdown of IB EE assessment criteria help you align your analysis with what examiners actually reward.
Step 1: Collect data that your research question actually needs
A common IB Sciences trap is collecting “interesting” data instead of necessary data.
If you’re generating primary data, design for:
-
Control: keep non-tested variables constant (or record them if they cannot be controlled).
-
Repeatability: enough trials to justify averages and variability.
-
Precision: instruments that match the scale of change you expect.
If primary data isn’t feasible, you can use secondary data, but you must be even stricter about relevance and limitations. This is where many EEs quietly lose credibility.
Helpful reference: Can I Use Primary Data in My Extended Essay?
Step 2: Organize raw and processed data so it’s impossible to misread
In IB Sciences, presentation is not decoration. It’s part of your thinking.
Build tables that:
-
separate raw vs processed data
-
include units in every heading
-
show uncertainties (absolute or percentage, depending on context)
-
use consistent significant figures tied to measurement precision
Then choose graphs like you’re choosing evidence for a debate:
-
scatter plots for relationships and trendlines
-
line graphs for continuous change over time
-
bar charts for discrete categories
Add error bars when they communicate something meaningful (variability or measurement uncertainty), not because you feel guilty.
If you want a strong model of what “analysis and line of argument” looks like, see RevisionDojo’s Criterion-focused guide: Criterion C: Analysis and Line of Argument.
Step 3: Process your data with methods that match the science
Processing is where IB Sciences students either look rigorous or improvisational.
Start with descriptive processing:
-
mean (or median if outliers are meaningful)
-
standard deviation, range, or uncertainty propagation
Then choose higher-level processing only when justified:
-
gradients from graphs (rates, constants)
-
derived values (e.g., concentration, energy change, index calculations)
-
regression and goodness-of-fit (when modeling is sensible)
The key is to show that your processing reduces noise and clarifies the relationship, rather than just adding complexity.
For a stats-focused companion that pairs well with this workflow, read: Using Statistical Analysis Effectively in an IB Science IA.
Step 4: Use statistics only when they answer a real question
Statistics in IB Sciences should do one of two jobs:
-
test whether an observed difference is likely meaningful
-
quantify the strength of a relationship
Examples:
-
t-test: comparing means between two conditions (with assumptions acknowledged)
-
chi-square: categorical outcomes or distributions
-
correlation/regression: relationships, with a clear statement about what correlation does and does not prove
What matters most is your justification. A single sentence like “a t-test was used” is not enough. You need to explain why the data type and design fit the test, and what the result implies for the research question.
Step 5: Interpret patterns like a scientist, not a narrator
This is the heart of data analysis in IB Sciences: interpretation.
Instead of “The graph shows an increase,” aim for:
-
trend: what changes, in what direction, and at what rate?
-
mechanism: why would science predict this pattern?
-
anomalies: what points don’t fit, and what are plausible reasons?
-
link back: how does this support, refine, or challenge your hypothesis?
A useful rule: every paragraph of analysis should contain at least one explicit reference to the research question. If it doesn’t, it may be interesting but not examinable.
Step 6: Evaluate reliability, validity, and limitations with specificity
Evaluation is not a list of generic errors. In IB Sciences, you earn trust through precision.
Discuss:
-
reliability: would repeated trials produce similar outcomes? (use your spread/SD as evidence)
-
validity: did you measure what you intended, or a proxy with hidden assumptions?
-
systematic vs random error: which dominates, and how does that shape confidence?
Then propose improvements that actually change the data quality:
-
more trials, better control method, recalibration, improved range of IV values
-
alternative measurement technique that reduces uncertainty
-
a redesigned method to reduce confounding variables
When you want to see what high-quality critique looks like in a real submission, browse an exemplar from the Coursework Library, such as this IB Chemistry Extended Essay Exemplar.
Step 7: Write the analysis so the examiner can follow your logic fast
Examiners don’t award marks for effort. They award marks for a clear line of reasoning.
In IB Sciences, that usually means:
-
state the claim (linked to RQ)
-
cite the relevant data (table/figure number)
-
explain what it means scientifically
-
qualify confidence (uncertainty, limitations, anomalies)
Avoid “data dumping” by making every figure earn its place. If a table doesn’t directly support the argument, it likely belongs in an appendix-style section (depending on your school’s guidance) or removed.
Final takeaway: turn IB Sciences data analysis into a repeatable system
In IB Sciences, the students who score well aren’t the ones with perfect data. They’re the ones who can explain imperfect data honestly, process it appropriately, and build a calm, logical argument from it.
If you want that same clarity in your exam prep and coursework workflow, RevisionDojo is built to support it: use the Questionbank to practice analysis-heavy questions, Study Notes to tighten scientific explanations, Flashcards to lock in definitions and methods, and AI Chat when a model or test stops making sense. When drafts need sharper criterion alignment, use the Grading tools, explore the Coursework Library, and bring in Tutors for targeted help.
Start small: pick one table or one graph from your EE, and run these seven steps on it today. That’s how IB Sciences analysis becomes something you can trust under pressure.