Can biased data still be useful in IB Math?

Yes, but only in a limited, honest way. Biased data can suggest a possible pattern worth investigating, especially if the effect is large and consistent. However, in IB Math you must avoid pretending it represents the entire population. The best exam responses separate “what the sample shows” from “what we can claim about the population.” You can still interpret the statistics, but you should add a limitation sentence that the sampling method may have excluded key groups. Examiners typically reward this restraint because it shows statistical maturity.

Does IB Math expect me to always identify bias?

Not always, but you are expected to identify it when it is present or strongly implied. In IB Math, sampling flaws are often embedded in the story: “online survey,” “asked students at lunch,” “gym members,” “volunteers,” and similar phrases. Questions may not say the word “biased,” but they give you enough to infer it. When you spot it, explain who is excluded and how that affects representativeness. That explanation is usually where interpretation marks live. Practising these prompts repeatedly is easier with a structured system like the RevisionDojo IB Math AI resource hub.

Is bias the same as having a small sample size?

No, and mixing them up costs marks. A small sample size mainly affects reliability because random variation is larger, so estimates can bounce around. Bias affects validity because the sample systematically misrepresents the population, even if the sample is huge. In IB Math, a large biased sample can be “precisely wrong,” which is actually more dangerous than being uncertain. Ideally, you comment on both when relevant: small sample size (weak evidence) and biased method (limited generalisation). If you want to see how IB treats size specifically, Why Sampling Method Matters More Than Sample Size makes the distinction clear.