[EM] Bias-Free, contd.

[MIKE OSSIPOFF]

Warren says that "cycle" is not a good name for an interval between two 
whole numbers of quotas. He hasn't said why. The function s(q) is a step 
function. Looking at it, you might wonder why Warren objects to "cycle".

By making cycles' s/q as close as possible to 1, BF eliminates bias, as 
defined by the starting definition that I stated, given a uniform frequency 
distribution. The distribution makes a little large-bias. Webster adds some 
of its own large-bias. It's obvious that Bias-Free will be less biased than 
Webster, the least biased of the traditional methods.

The apportionment test of BF that I referred to was done by Dan Bishop, for 
2000 & 1990.

Mike Ossipoff

_________________________________________________________________
Dave vs. Carl: The Insignificant Championship Series.  Who will win? 
http://clk.atdmt.com/MSN/go/msnnkwsp0070000001msn/direct/01/?href=http://davevscarl.spaces.live.com/?icid=T001MSN38C07001