I suppose that the concept of a "population" versus a "samples from a
population" were formulated to deal with problems in fields like
sociology, medicine or advertising. For example, all 125,641 US male
drivers, in a given age group, can be defined as a population. It is
not practical to study each individual. In order to learn something
about the entire population statisticians may randomly select a sample
of 10,000 people and study them. For example, they may collect
information on health, income, education, political prefernces, race,
sexual orientation, attitude toward the war etc, etc. On that basis
they make predictions about the entire population.
What should be called a population when students measure gravitational
acceleration? Each set of five results can be called a sample. Students
try to estimate the true g in their location. Is this their
"population' ? If not then what is the population? Why do we need the
concept of population when true values, or differences between true
values, are measured with instruments of limited precision? Individuals
can be considered as parts of instruments.
Ludwik Kowalski
Let the perfect not be the enemy of the good.