# Data Types

Basically the data is divided into continuous and discreet. The first is defined as any value between two limits any, such as a diameter. So it is a value to be "broken". It is continuous data, issues that involve age, income, spending, sales, billing, among many others.

When speaking of discrete values, one approaches an exact value, such as quantity of defective parts. This type of variable is commonly used to address child numbers, satisfaction and overall nominal scales.

The typology of the data determines the variable, so it will be continuous or discrete. This means that by defining a variable with continuous or discrete, in the future it has already been defined what kind of treatment will be given to it.

According to what we said earlier, a statistical analysis essentially distinguishes two phases:

A first phase in which we seek to describe and study the sample (Descriptive Statistics) and a second phase in which we seek to draw conclusions for the population (Inductive Statistics).

Phase 1 (Descriptive Statistics): we try to describe the sample, highlighting the main characteristics and properties.

Phase 2 (Inductive Statistics)Some known properties (obtained from a descriptive analysis of the sample), expressed through propositions, are imagined as more general propositions that express the existence of laws (in the population).

However, contrary to the inferred propositions, we cannot say that they are false or true, since they have been verified over a restricted set of individuals, and therefore are not false, but have not been verified for all individuals in the Population, so neither we can say that they are true.

There is thus a certain degree of uncertainty (percentage of error) that is measured in terms of probability.

Considering what was said earlier about Inductive Statistics, we need here the notion of Probability to measure the degree of uncertainty that exists when we draw a conclusion for the population from the observation of the sample.