Reacción a "Study finds three sigma probability of biological activity outside our solar system"

Most scientific experiments have some element of uncertainty. This could be sampling uncertainty – maybe we only have a small number of observations. Or it could be measurement error – maybe our measurements are noisy. If we picked 5 random men and 5 random women from the street, sometimes we will find that the men are taller on average than the women, but occasionally we will find that the women are taller on average than the men. If we want to conclude that men are typically taller than women, we need to collect enough data to be confident that the differences we observe are genuine differences, and not just chance fluctuations. The more data that we collect, the more certain we can be of this. “Three-sigma” is a threshold saying that differences observed in the experiment are sufficiently notable that we can exclude the possibility of a chance finding except in rare cases – equivalent in rarity to tossing a coin 10 times and getting the same result each time. “Five-sigma” is a stricter threshold – equivalent to tossing a coin 20 times in a row and getting the same result each time. It’s still possible that we were simply lucky – and the more data that we look at, the greater the chances of making an observation that is purely a chance finding. But a five-sigma finding is one that would only arise purely by chance exceptionally rarely, and so we can be very confident that this observation isn’t just a chance finding. A separate question to uncertainty is bias – it is possible that there is some flaw with the experiment. This is not something that can be ruled out by statistics. A “five-sigma” finding is therefore exceptionally unlikely to arise due to chance alone: it is either a true result or an experimental error.