IB Mathematics Internal Assessment
Own data is the best gift you can give your exploration. It is also where a lot of students quietly lose marks, because data collected in a hurry cannot bear the weight of the mathematics built on top of it. Here is how to collect it so it survives the questions a reader will ask.
Data you gathered yourself gives you three things that a downloaded spreadsheet cannot. You can explain exactly where every number came from. You have real decisions to justify, about what to measure and how. And you get real mess to deal with, which is where reflection comes from. Students who use a tidy public dataset often struggle to find anything to say about it. Students who timed their own laps, counted their own customers or logged their own sleep never run out.
The catch is that own data is only worth having if it is collected with some care. A reader is not marking your enthusiasm. They are asking whether the mathematics you did is meaningful, and that depends on whether the data underneath it means anything.
The most common mistake happens before any data exists. A student collects a pile of measurements and then goes looking for something to do with them. The exploration ends up shaped by whatever the data happens to allow, and the aim is vague.
Reverse it. Write your question first, then work out what you would have to measure to answer it, and only then collect. Before you record anything, be able to say in one sentence what each variable is for. If you cannot say what a column is for, do not collect it.
Hours studied, hours slept, mood, coffee, screen time, all logged daily. There is no question driving it, so nothing tells you which of the dozen possible relationships matters, and whichever one you pick will look arbitrary.
The question dictates the data: one repeatable task, one recorded time, one clearly logged time of day. Everything you collect has a job to do, and everything you leave out is a decision you can defend.
Data holds up when every value was produced in the same way. Before you start, decide and write down your method: what counts as one observation, what units you use, how you measure, and to what precision. A stopwatch read to the nearest tenth of a second in one session and the nearest second in another gives you numbers that cannot honestly be mixed.
If your data comes from people, the biggest threat is a sample that does not represent the group you are talking about. Surveying your own friend group about something and then drawing conclusions about teenagers in general is the classic version. The mathematics may be flawless and the conclusion still unsupported.
You do not need a perfect sample. You need to know what your sample is, say so plainly, and limit your claims to match. "Among the students I surveyed in my own year group" is a claim you can defend. Anything wider needs a reason.
Ask yourself these before you collect:
Real data has missing values, obvious slips, odd outliers and days when the method went wrong. This is not a failure. It is material. What matters is what you do about it and whether you say so.
If a value looks wrong, find out why before you remove it. Removing a point because it spoils your line of best fit is not cleaning, it is choosing your result. Removing a point because you know the stopwatch was started late is a decision you can explain. The difference is a sentence or two of reasoning, and that reasoning is exactly the kind of thinking a reader is looking for.
You notice one value far from the rest, check your log, find you recorded it during a different task, and say so. You then show what the analysis looks like with and without it. The reader sees a person in control of their data.
The data goes in with twenty values and comes out with nineteen, and nothing is said. If the reader spots it, everything else you have done becomes harder to trust.
Students often stop collecting when they get bored, or when the deadline arrives. Decide your stopping point in advance, based on what your mathematics needs. If you plan to fit a model, ask how many points would let you check that model against data it was not built from. Collecting a little extra, kept aside to test your model, is one of the most persuasive things you can do.
Also collect early. Data is the one part of the exploration you cannot speed up at the last minute, because the world will not repeat a week on demand. If your first batch turns out to be unusable, you want time to collect it again.
Every dataset has limits, and naming them precisely is a strength. Vague lines such as "the data may not be perfect" earn nothing. Specific ones do: which conditions varied, how the sample was chosen, what your measuring method could and could not detect, and how each of those might have affected your result. Then say what you would change with more time. That is reflection with substance, and it grows naturally out of data you collected yourself.
Planning to collect your own data and not sure it will stand up?
Send me your question and how you plan to gather the data. I read real submitted explorations every year, and I will tell you honestly where the weak points are while there is still time to fix them.
Get the IA sorted with meMore like this: all Maths IA guides. Related: choosing a Maths AI SL IA topic when you do not like statistics
A new IA guide goes up most nights. If you would rather not keep checking, leave an email and I will send the one that matters that week. No selling, and leave whenever you like.
If you are under 18, use a parent's email. I do not correspond privately with students.
These pages are free and stay free, but they are general and your IA is not. Send me your research question, or whatever exists so far, and I will tell you in writing whether the topic has a ceiling on it, where the marks are going, and what to change first. That costs nothing and it comes back within 24 hours.
Written by a serving IB Diploma and Career-related Programme Coordinator and Head of Mathematics, who reads internal assessments across every subject group every year. If you then want the whole draft reviewed properly against all five criteria, that is the paid one, and it is refunded if it does not name at least three specific things to fix.
Get a free verdict Full written review, $99
I never write any part of it. Not a sentence, not a calculation, not your data. Under 18: a parent buys this and the thread is with them. I do not work with students at my own school.