We explain what a statistical population is, its characteristics and what types exist. Also, what is a statistical subpopulation.
What is a statistical population?
A statistical population (or, in an area clearly and openly referred to the world of statistics, simply as a population), is the set of elements that are of interest for an experiment, a study or a consideration of some kind. The elements that compose it can be, for example, individuals, animals, phenomena or events.
In fact, all forms of statistical study aspire to provide information about a population determined in advance, whether real and existing (such as the total number of voters in a country), or hypothetical (such as the total number of times we toss a coin in the air). ).
Therefore, the statistical population represents the universe of the elements to be considered, that is, its full totality, and in this it is distinguished from a statistical sample. The latter is a portion of said universe, that is, a subset of the population, which is taken for analysis since it is much smaller and more manageable than the total, but still representative of it.
Statistical samples are studied to obtain probable conclusions regarding statistical populations whose individual and detailed study would be practically impossible.
For example, if a jar contains 50 five-cent coins and another 50 ten-cent coins, the statistical population will be 100 coins, since when you reach in and take out a sample, there will be that total number of elements, among which to take a handful .
See also: Inferential statistics
Characteristics of a statistical population
A statistical population is characterized by the following:
- It constitutes a total of elements of statistical interest for some reason, from which representative samples can be taken.
- It can be more or less uniform or heterogeneous and in the same way it can be constituted by real or imaginary, finite or virtually unlimited elements.
- Not to be confused with statistical sample.
Types of statistical populations
Statistical populations are classified into two, according to their finitude:
- Finite statistical population. As its name indicates, it is made up of a delimited and encompassable quantity of elements, which at a given moment in time is equivalent to a specific number. For example: the number of cars in circulation in a city on a Monday morning.
- Infinite statistical population. On the other hand, this type of statistical populations have a virtually unlimited number of elements, that is, they do not have a specific purpose at a given moment, either because they are really unlimited, or because their number is so large that we could never know with certainty. For example: the number of sodium atoms in the universe.
Statistical subpopulation
A statistical subpopulation is a portion of the statistical population which can be considered a universe in itself, given that Its members share an exclusive trait with respect to the rest. That is, a statistical subpopulation is a population within the population, which is generated by adding specific traits to the selection criterion.
For example: in the universe of current European citizens, it is possible to choose different subpopulations according to their particular nationality: Italian, French, Spanish, German, etc. Within each of these subpopulations, it is possible to do the same thing again if we consider the subpopulations of German men and German women, for example.
Continue with: Probability and Statistics
References
- “Statistical population” in Wikipedia.
- “What is population?” at National Institute of Statistics (Chile).
- “Population and sample” at the National Institute of Educational Technologies and Teacher Training (Spain).
- “Statistical Population” in OECD Glossary of Statistical Terms.
