How to write the IB Sports, Exercise & Health Science IA

The Sports, Exercise & Health Science IA is the one piece of coursework your SEHS grade is marked on internally — worth 20% of your final grade at both SL and HL. What makes it different from a Chemistry or Physics IA is that your data comes from people: real participants whose hearts beat at different rates, who warm up differently, and who simply vary from one another. That human element is where the marks are won and lost. Most students lose marks not because they cannot understand the physiology, but because they choose a vague question, never standardise their conditions, or stop at a table of means without ever running a statistical test. This guide walks you through the whole thing: what the IA is, how it is marked, exactly how to write each part, and what separates a top-band investigation from an average one.

It helps to be honest about why the SEHS IA feels harder than it looks. A chemistry investigation lets you control almost everything: the same reagents, the same glassware, the same temperature each time. With human participants you are handed variation you did not ask for and cannot remove. Two people can complete an identical step test and produce recovery heart rates twenty beats apart simply because one trains five times a week and the other does not. The skill the IA is really testing is whether you can design around that variation rather than pretend it is not there — choosing participants thoughtfully, holding conditions constant, repeating measurements, and then using statistics to separate a real effect from background noise. Once you see the whole task through that lens, the four criteria stop feeling like arbitrary hoops and start looking like a sensible checklist for doing fair science on people.

The IB SEHS IA at a glance

Under the current syllabus the SEHS IA is called the Scientific Investigation: a single, focused, individual investigation reported concisely — typically 6–12 pages within the 3,000-word guideline. It is marked out of 24 across four equally weighted criteria, and the same criteria are shared with Biology, Chemistry, Physics and Environmental Systems & Societies. There is no separate mark for "personal engagement"; your engagement should show through the relevance and ownership of an investigation that genuinely interests you — a sport you play, a training method you use, or a health question you care about. Because your participants are human, the design also has to satisfy the ethical expectations that come with testing people: screening, consent and welfare are not optional extras, they are part of Research design.

How the SEHS IA is marked: the four criteria

Every mark comes from one of these four criteria, each worth 6. The single most useful habit you can build is to stop thinking of the IA as an essay and start thinking of it as four separate scoring opportunities. Examiners do not read your report from start to finish and award a holistic impression; they move through each criterion in turn, looking for the specific features described in the band descriptors, and they tick them off as they find them. If a feature is missing — a measure of spread, a justified hypothesis, a referenced comparison — the mark for that criterion is capped regardless of how polished the rest of the writing is. So write your IA criterion by criterion, and for each one ask the same blunt question: have I clearly shown the examiner the thing this criterion rewards? Here is what each one is looking for:

Notice how the four criteria form a chain. A weak research question makes a convincing conclusion almost impossible, because there is nothing precise to conclude about. Sloppy standardisation in the design resurfaces in the evaluation as a limitation you cannot easily dismiss. And data that never makes it past a table of means leaves your conclusion with nothing to stand on. The students who score well treat the design stage as where most of the marks are actually decided: an hour spent narrowing the question, pinning down what will be held constant, and checking that the measurement is genuinely quantitative will save far more marks than any amount of careful prose later. Get the front end right and the back end tends to write itself.

How to write a SEHS IA, step by step

1. Write a research question on human participants. Build the investigation on a measurable physiological response, with one variable you can change over a range and one you can measure — for example, how recovery heart rate depends on stepping rate.
2. State a testable hypothesis. Predict the direction of the relationship and explain why, drawing on the exercise physiology rather than guessing.
3. Plan sampling and standardisation. Decide how participants are chosen and spell out how every condition — warm-up, rest, timing, technique — is held the same across all of them.
4. Complete PAR-Q and consent. Screen each participant with a PAR-Q, take written informed consent, and record how their welfare and personal data are protected.
5. Write a reproducible method. Describe the protocol so precisely that another student could repeat it identically, including apparatus, timings and exactly how each reading is taken.
6. Collect raw data across repeats. At least five values of the independent variable across a sensible range, with repeated trials so inter-participant variation is captured, recorded with units.
7. Process with mean ± SD and a test. Calculate the mean and standard deviation for each condition, then run an appropriate statistical test and state the decision it leads to.
8. Plot a graph with error bars. Plot the means with error bars showing the standard deviation, so the spread between participants is visible at a glance.
9. Write a justified conclusion. Answer the question, justify it against the physiology, and compare it to referenced secondary data.
10. Evaluate honestly. Weigh limitations by impact — especially participant variability and standardisation — and propose specific improvements and an extension.

A few of these steps deserve a closer look, because they are where SEHS investigations most often come apart. Sampling and standardisation, step three, is the one students underestimate most. Because your participants are people rather than identical samples of a chemical, every uncontrolled difference between them — fitness, sleep, caffeine, time of day, how vigorously they swing their arms during a step test — becomes noise in your data. You cannot eliminate that variation, but you can decide which parts of it to control and say so explicitly. Recruiting from a single training squad, fixing the warm-up, and testing everyone at the same time of day are not bureaucratic box-ticking; they are the difference between error bars that overlap and error bars that separate. The PAR-Q and consent step is equally non-negotiable: it is assessed within Research design, and an investigation on human participants that skips it has a ceiling on its mark no matter how elegant the rest is.

Step seven — processing into a mean and standard deviation and then running a statistical test — is the other classic stumbling block. The standard deviation is what tells the reader how much your participants varied around each mean, and the statistical test is what tells them whether the difference you observed is large enough to take seriously rather than a fluke of a small sample. Decide on your test before you collect data, because the test you intend to run shapes how many participants and conditions you need. A paired comparison of two conditions, a correlation across a gradient, and a comparison of several groups each point to a different test, and choosing the right one — then stating the decision it produces in plain language — is exactly what the Data analysis criterion rewards.

SEHS IA structure: what goes in each section

There is no single mandated layout, but the clearest structure that maps onto the criteria is:

• Research question & hypothesis — the question on human participants, the physiology behind it, and a testable prediction.
• Variables — independent (range + units), dependent (how measured), and the conditions standardised across participants.
• Ethics & screening — PAR-Q health screening, written informed consent, and how welfare and data are protected.
• Method — apparatus, sampling and a reproducible procedure another student could repeat.
• Raw data — a clear table with units, including every participant and repeat.
• Data processing — mean ± standard deviation per condition, with sample calculations.
• Statistical test — an appropriate test, the value obtained, and the decision it leads to.
• Graph — processed means with error bars showing the standard deviation.
• Conclusion — answer, physiological justification, comparison to secondary data.
• Evaluation — limitations by impact, improvements, extension.
• References — a consistent citation style throughout.

You do not have to use these exact headings, and you should not pad the report to hit them all if a section genuinely has nothing to say. The point of the structure is that it forces every criterion to surface somewhere obvious, so an examiner skimming for the features they need can find each one quickly. A common and avoidable mistake is to bury the standardisation in the middle of the method, or to mention the comparison with published data only in passing in the conclusion. Give each rewarded feature its own clearly signposted home. Within the 3,000-word guideline, that discipline also keeps you concise: a report that knows what each section is for rarely runs long, while one that wanders tends both to exceed the limit and to leave the examiner hunting for marks that are technically present but hard to see.

What a strong vs weak SEHS IA looks like

The fastest way to lift your marks is to see the difference. Each pair below shows the same piece of work written two ways — the version that quietly loses marks, and the version that earns them. As you read, notice that the strong column is rarely longer or more sophisticated; it is simply more specific. It names things, attaches numbers and units, and makes the reasoning visible instead of leaving the examiner to assume it.

The research question

Statistics and processing

Evaluation

Read those three pairs again and you will see the same move each time. The weak version describes; the strong version measures, names and reasons. "Higher at faster stepping rates" becomes "128 ± 9 bpm versus 104 ± 7 bpm, p = 0.003". "Test more people" becomes "the wide error bars point to baseline fitness, so standardise on one squad". You are not being asked to write more — you are being asked to write more precisely, and to make the evidence and the reasoning behind every claim impossible to miss. If you can rewrite your own draft so that every sentence in the conclusion and evaluation does what the strong column does, you are most of the way to the top band.

Common mistakes that cost marks

• No ethics or consent. Testing people without a PAR-Q and written informed consent is a serious Research design failure, not a formality.
• No standardisation. If warm-ups, rest periods or technique differ between participants, the comparison is no longer fair and the data is hard to defend.
• No statistical test. Means with no standard deviation and no test cap the Data analysis criterion — you must show whether a difference is real.
• Ignoring inter-participant variation. People vary; a graph with no error bars hides the spread that drives your evaluation.
• No physiological justification. A conclusion that never explains the mechanism, or never cites secondary data, cannot reach the top band.
• "Human error." Examiners read this as "I didn't analyse my method." Always name a specific, real weakness such as an unstandardised warm-up.
• Going over 3,000 words. Examiners stop reading at the limit — be concise.

SEHS IA — frequently asked questions

Guidance written by experienced IB examiners and aligned to the current Sports, Exercise & Health Science guide. Not affiliated with or endorsed by the International Baccalaureate Organization.