IB Maths AI: Interpreting Real-World Data FAQ

Common pattern I see every mock season: a student finds r = 0.87, writes 'strong positive correlation' and stops. That earns the calculation mark but not the context mark, because the examiner needs to see what's correlated — e.g. 'strong positive correlation between hours studied and exam score.'

3 things to check before you write your final sentence:

1. Have I named the actual variables, not just 'x and y'?
2. Have I included the correct units (kg, USD, years)?
3. Have I answered what the question actually asked — not just what I calculated?

Rounding is the sneaky one. If you round a gradient to 2 s.f. mid-calculation and then use that rounded figure in a later step, the final answer can drift outside the accepted range — examiners call this premature rounding, and the IB's mark schemes explicitly penalise it. Keep at least 3-4 significant figures until the final line, then round only your final answer to the accuracy the question asks for.

Quick tip: after every statistical or modelling answer, ask yourself 'would this sentence make sense to someone who's never seen the question?' If the answer is no, you haven't finished interpreting the data yet.

Worked example: A study gives r = -0.62 for hours of screen time vs sleep duration.

Weak/poor answer: 'r = -0.62, negative correlation.'

Full-mark answer: 'There is a moderate, negative correlation between hours of screen time and sleep duration (r = -0.62), suggesting that students who spend more time on screens tend to sleep less. This does not prove screen time causes reduced sleep, since other factors may be involved.'

That's three sentences, three separate marking points.

Step-by-step:

1. Calculate χ² and degrees of freedom.
2. Compare to the critical value at the given significance level (commonly 5%).
3. State the decision: reject or do not reject H₀.
4. Write the conclusion using the actual variables from the question, not 'H₀' or 'the variables.'

Examiners frequently deduct the final mark when step 4 is skipped, even if steps 1-3 are flawless.

Common mistake: rounding a mean or standard deviation to 2 d.p. early, then using that rounded figure to calculate a z-score or confidence interval later. The final answer can land just outside the accepted range, costing an accuracy mark that has nothing to do with your method.

Quick tip: read five or six official mark schemes back-to-back and note the exact phrasing examiners accept for 'weak/moderate/strong' and 'positive/negative' correlation — reusing that exact vocabulary in your own answers removes any ambiguity for the marker.

According to the IB's Mathematics: analysis and approaches and Mathematics: applications and interpretation guide (first assessed 2021, current for the 2025 syllabus cycle), IA Criterion E — 'Interpretation' — is worth 6 of the 20 IA marks, making it the single largest criterion. SL students don't sit Paper 3, so their data-interpretation demands sit mainly in Papers 1 and 2 and the IA.

To hit the top band (5-6 marks), your commentary needs to discuss the validity and limitations of your model or method, not just restate the numeric result. A student who calculates a regression equation but never questions whether a linear model actually suits the data will cap out well below full marks on this criterion alone.

AI suits students heading toward social sciences, business, psychology or economics degrees where statistical literacy matters more than abstract proof. AA suits those heading toward engineering, physics or maths-heavy degrees. Either route, the interpretation skill — writing a conclusion in context — is something worksheets and past-paper practice on revisionprep.com can build steadily over a term rather than needing to be perfect from day one.

Before spending on anything, check exactly where marks are being lost — is it the calculation, the rounding, or the written conclusion? For most students it's the last one, which needs targeted practice against real mark schemes, not more generic maths tuition.