Description of One Variable
Measures of Distribution
The values taken from the source can be grouped on brackets, the number  of times that each brackets ocurrs constitute the variable's frequency distribution.
Histograms
In a histogram the relative frequency of each bracket is ploted on a bar graph. There are several important characteristics about a histogram:

Mode:is the distribution that corresponds to the highest bar in the histogram. It signifies the most frequent bracket observed.
if the bars move away from the mode in a histogram in either direction; the distribution is called unimodal, on the other hand if there are many other peaks is called multimodal, and if there are 2 peaks is called bimodal.
Symmetry: A distribution is symmetruc when the bars on the histogram mirrors the bars on the left.
Skewness: these are the distributions that are not symmetric

Cumugrams
These are plots of cumulative distributions, they show the fraction of values that are less than or equal to some particular value
Measures of Centrality
We always wnats to specify  a measure that is typical of the value, and that is called measure of centrality. The most common measures of centrality are  the median, mean and the mode.
The mediam is the medium value, while the mean is the average of its values.
Measures of Locations
Fractiles- serve as a measure of location. They represent the lowest value of a variable below which some specified fraction of the observations lie. The median is equal to the .5 fractile.
Measures of Spread
In analyzing the observations, we are often interested on knowing how the values of the variable spread out. Several measures report the spread or dispersion of a variable's value.

Standard deviation-it measures the dispersion of values around the mean of a variable. In general the closer the oobservations are to the mean, the smaller the standard deviation accentuates values that are far from the mean.
Range- It simple measures the difference between the highest and lowest value of the variable, it should be used with cautions because it is strongly affected by the outliers.
Interquartile Range-measures the difference between fractiles that contain half of the observations.

Types of Variables
Ratio Scale Variables- if its values have no natural upper bound and can not be negative.
Difference-scale variables- it can be either negative or positive with no natural upper or lower bound.
Ordinal Variables- are also numeric, because the numerical differences between ordinal variables are not equal the mean and standard deviation are, strictly speaking innapropiate measures
Categorical Values- consist in qualitative variables such as religion or marital status.
Dummy Variables- such as female or male which will represent a value of 0 or 1.

Return to Index