A statistical proposition attributes a quantitative, numerical
property to some class of things.
Example:
The population of South Carolina is approximately 3 million people.
The homicide rate in the United States is 9.6 per 100,000.
Ninety-seven percent of the people who eat chocolate bunnies
start with the ears.
A statistic is no better than the reasoning process by which it
was derived. When the reasoning is fallacious, arbitrary, or
flawed in some other way, the statistic may bear no relation to
reality.
Statistics deal with classes of things, and for the sake of
continuity with previous chapters, we use the letter S to
stand for a class (and T, U, etc., if there is more than one
class). Statistics also deal with the properties that members of
a class possess, especially the measurable properties.
The genus-species relation applies to properties as well as to
classes of objects.
In statistics, a generic quality is called a variable, and the
specific qualities are called values of the variable.
Example:
Red and green are values of the variable Color.
A, B, AB, and O are values of the variable Blood type.
Republican and Democrat are values of the variable Political party affiliation.