The book is, therefore, highly recommended and would
serve as useful background reading and support for any useful course in
reasoning ability and scientific understanding. I have even purchased
a few copies and sent them to people I think can profit from the book.
“Throughout history, the world has abounded with differences that are
today called "disparities" or "inequities," even in situations
where they cannot be explained by discrimination. At one time, in czarist
Russia, nearly all of the members of the St. Petersburg Academy of Sciences
were of German ancestry, even though people of German ancestry were only about
one percent of the population of Russia. Today, more than 40 percent of all
the billionaires in the world are in one country--the United States. The list
could go on and on, until it filled a book. But, however common such statistical
disparities have been around the world and throughout history, many continue
to reason as if any statistical differences between any groups and suspicious,
if not sinister.
“Another fallacy, already noted in Chapters 5 and 7, is what might
be called the fallacy of changing composition. When statistical categories
are, compared over time, the changing relationships among these categories
can be completely misleading as to what is happening to the people or the
nations in those categories, when the composition of these categories is changing
over time. There may be growing inequalities between those categories during
the very same span of years when there is a lessening of inequality
between the people or nations who constitute those categories. Moreover, important
conclusions and decisions can be based on this fallacy.
“For example, the growth of international free trade has been said
to increase inequality among nations because the 23-to-one ratio between the
twenty richest and twenty poorest nations in 1960 rose to a 36-to-one ratio
in 2000. But the nations constituting the 20 richest and 20 poorest were different
in 1960 and 2000. Comparing the same twenty richest and twenty poorest
nations of 1960 over those decades shows that the ratio between the richest
and poorest declined to less than ten-to-one. This leads to the directly
opposite conclusion, suggesting that freer international trade may have helped
reduce inequalities among nations, allowing some of the initially poorest
to rise out of the category of the bottom twenty.
“Whatever the reason for the declining inequality, the fallacy of believing
that international inequality had increased, when in fact it had decreased,
is similar to that in an old joke about automobile accidents in Manhattan.
In this joke, one friend says to another that statistics show that a man is
hit by .a car in Manhattan once every 20 minutes. To which the other replies,
"He must get awfully tired of that." The fallacy here is that it
is obviously not the same man each time. The very same fallacy underlies much
more serious conclusions about both personal and international inequalities
over time, when it is not the same individuals or the same nations that are
being compared, since each moves from one category to another over time. The
changing composition of the categories makes conclusions based on comparisons
between the categories fallacious.
“Statistics are no better than the methods and definitions used in
collecting them. Without scrutinizing those methods and definitions, we cannot
assume that comparable people are being compared, whether comparing the incomes
of high school dropouts with college graduates, the incomes of members of
different ethnic groups who have the "same" education, or the incomes
of single women with married women, when "single" women includes
women who were married for years before getting divorced. Nor can statistics
about the amount of air pollution in populated areas versus open space tell
us anything about whether letting people move into unpopulated areas will
increase the total pollution over all, since it is people--not their locations--that
generate pollution.
“Perhaps most dangerous of all is the practice of not subjecting fashionable
beliefs to the test of facts, but instead accepting or rejecting beliefs according
to how well they fit some pre-existing vision of the world. The idea that
government intervention is needed to create "affordable housing"
is an idea that makes sense only in the context of a preconceived notion,
while mountains of hard evidence point in the exact opposite direction. The
belief that ghetto riots such as those of the 1960s are a reaction against
poverty, discrimination, unemployment, and blighted communities simply will
not stand up in the face of hard evidence of when and where those riots took
place, which were not in the places or times where these factors were worse.
“The entire educational and employment history of women in the first
half of the twentieth century is almost invariably ignored, even in scholarly
studies, to concentrate attention on what has happened since 1960, which can
be made to fit a preconceived vision of the reasons for women's rise. Similarly
with blacks, whose rises out of poverty and into middle class occupations
are likewise traced almost invariably from some point after 1960, and attributed
to the civil rights movement and government actions of that decade, even though
the most dramatic rises of blacks out of poverty occurred in the two decades
1960. Nothing is more fallacious than to ignore a trend that began
years before some policy or action that is credited with whatever happened
as a continuation of a pre-existing trend. Similar fallacies have appeared
in discussions of things ranging from automobile fatality rates to market
shares of companies after an antitrust lawsuit.” [Pages 217-219]
