orb@whuts.UUCP (SEVENER) (10/31/85)
> > [Richard Carnes, ihnp4!gargoyle!carnes, in top note of this sequence] > > I would also argue, as indeed I have already, that such countries as > > China, Cuba, and Nicaragua have made giant strides in reducing hunger > > in their countries, mainly because of policies that redistribute > > power over food-producing resources in the direction of more > > equality. > > The following is from "China, Alive in a Bitter Sea", by Fox Butterfield, > Bantam Books, p. 15. > > > > For recent Western Studies show that food consumption per capita > > > is actually only about what it was in the mid-1950s, and, more > > > surprisingly, no better than in the 1930s, before World War Two. > > > > > > These studies suggest that the average daily calorie supply in > > > China is between 2,000 and 2,100 per person. Two thousand > > > calories a day is the level of India, 2,100 is the norm in Pakis- > > > tan. Americans eat an average of 3,240 calories a day. > Jan Wasilewsky A common statistical fallacy is to cite overall averages or per capita figures. Such figures show absolutely *nothing* about *distribution* of a given good, be it income, food or otherwise. For example, ten people could have a thousand dollars in total, with one person having $991 and the rest having $1 apiece. The per capita dollars would be $100, but the median would be $1 - an enormous difference. On the other hand the same ten people could have only $500 yet if this were divided equally they would each have $50 apiece. Their per capita dollars would be only $50 but their median dollars would also be $50. It seems obvious that more people are better off in the latter more equal distribution. Yet if one only compares the per capita figures of $100 (when 9 of the people actually only have $1) vs. $50 it would seem the first group is better off. Thus per capita figures are very misleading and almost never used by social scientists doing serious comparisons of goods distributions. Instead the more appropriate figure is the *median* i.e. the cutting point at which half the people get more than the given amount and half the people get less than the given amount. Even the median however is not a good measure of the overall distribution or its inequality. For that one can calculate a gini coefficient which shows the difference between the actual distribution and one of total equality. The point is that Jan has offered absolutely no evidence to refute Richard's claims about the *distribution* of food in Communist countries. tim sevener whuxn!orb