How can I teach sampling bias with meteorite falls and finds?

Question

Accepted Answer

Compare the types of meteorites among falls and finds to show how the way data are collected affects results, a key idea in statistics and science.

Meteorite statistics provide a vivid example of sampling bias.

The data. Among falls, meteorites seen to fall and then collected, about 4 to 5 per cent are iron meteorites and about 80 per cent are ordinary chondrites. Among finds outside deserts and Antarctica, iron meteorites are far more common, making up a large share of finds in some regions.

The question. Ask pupils: why might iron meteorites be so common among finds when they are rare among falls?

Explanations. - Recognition. Iron meteorites look unusual, are very heavy and attract magnets, so people notice them and keep them. Stony meteorites look like ordinary rocks. - Survival. Iron meteorites resist weathering longer than many stony meteorites, especially fragile carbonaceous ones, so they survive to be found. - Detection. Metal detectors find irons easily.

Other biases. Antarctic and desert collections have their own biases: they favour meteorites that survive long periods and stand out against ice or sand. Witnessed falls are biased towards populated areas and daytime or evening events.

Activity. Give groups a 'population' of coloured counters representing different meteorite types, with most being one colour. Then apply rules: some colours are more likely to be 'found' or to 'survive'. Pupils compare the sample with the original population.

Wider lessons. Discuss sampling bias in other fields, such as surveys, medical trials and fossil records.

This helps pupils appreciate that how data are gathered shapes the conclusions.

How can I teach sampling bias with meteorite falls and finds?

More about for teachers