I call heads....
Posted by Richard Linney on Jan 08, 2002 at 05:16
Euro coin accused of unfair flipping
The introduction of the Euro, the largest currency switch in history, has proceeded with few problems - until now. Polish statisticians say the one Euro coin, at least in Belgium, does not have an equal chance of landing "heads" or "tails". They allege that, when spun on a smooth surface, the coin comes up heads more often.
The observation is not to be taken lightly on a sports-mad continent where important decisions can turn on the flip of a coin. But the accusation of bias has been countered by statistical analysis from, of all places, Euro-sceptic Britain. The UK is one of only three EU countries that have not adopted the common currency.
Tomasz Gliszczynski and Waclaw Zawadowski, statistics teachers at the Akademia Podlaska in Siedlce, received Belgian Euro coins from Poles returning from jobs in Belgium and immediately set their students spinning them. Gliszczynski says spinning is a more sensitive way of revealing if a coin is weighted than the more usual method of tossing in the air.
The students had already spun the Polish two-zloty piece more than 10,000 times to show it was biased. But for the Belgian Euro, they have so far managed only 250 spins.
Of these, 140, or 56.0 per cent, came up heads. Glyszczynski attributes such assymetry to a heavier embossed image on one side of the coin. All Euros have a national image on the "heads" side and a common design on the "tails". Belgium portrays its portly king, Albert, on the heads side.
But Howard Grubb, an applied statistician at the University of Reading, notes that, "with a sample of only 250, anything between 43.8 per cent and 56.2 per cent on one side or the other cannot be said to be biased".
This is because random variation can produce such scatter even if the coin is truly unbiased. With a larger number of spins, such randomness would even out and results would approach 50:50.
The range of 6.2 per cent on either side of 50 per cent is expected to cover the results, even with a fair coin, in 95 of every 100 experiments. Nonetheless, Grubb cautions, the Polish result is at the outside of this range, and would be expected in only about 7 of every 100 experiments with a fair coin, leaving a glimmer of hope for their hypothesis. Clearly, more research is needed.
Gliszczynski plans to continue his experiments - aimed mainly at teaching his students statistics - with the German Euro, which has an eagle on its heads side, and present them at a conference in February.
New Scientist carried out its own experiments with the Belgian Euro in its Brussels office. Heads came up five per cent less often than tails. This looks like the opposite of the Polish result but in fact - in terms of statistical significance - it is the same one.
Article available here.
17:39 04 January 02
Post a Followup