- Descriptive statistics
- Categories
- Inferential statistics
- Categories
- Differences between descriptive and inferential statistics
- References
The descriptive and inferential statistics are part of the two main branches where the statistic is divided, the exact science that is responsible for extracting information on several variables, measuring them, controlling and communicating in case there is uncertainty.
In this way, statistics aims to quantify and control both scientific and social behaviors and events.
Descriptive statistics is responsible for summarizing the information derived from the data relating to a population or sample. Its objective is to synthesize this information in a precise, simple, clear and orderly way (Santillán, 2016).
This is how descriptive statistics can indicate the most representative elements of a group of data, known as statistical data. In short, this type of statistic is responsible for making descriptions of said data.
For its part, inferential statistics is responsible for making inferences about the data collected. It throws conclusions different from what is shown by the data itself.
This type of statistics goes beyond the simple compilation of information, relating each piece of information to phenomena that can alter its behavior.
Inferential statistics reach relevant conclusions about a population from analysis of a sample. Therefore, you should always calculate a margin of error within your conclusions.
Descriptive statistics
It is the most popular and well-known branch of statistics. Its main objective is to analyze variables and subsequently describe the results obtained from said analysis.
Descriptive statistics seeks to describe a group of data in order to pinpoint precisely the characteristics that define said group (Fortun, 2012).
It can be said that this branch of statistics is responsible for ordering, summarizing and classifying the data resulting from the analysis of the information derived from a group.
Some examples of descriptive statistics might include the population censuses of a country in a given year or the number of people who were admitted to a hospital within a given time frame.
Categories
There are certain concepts and categories that are exclusively part of the field of descriptive statistics. Some are listed below:
- Dispersion: it is the difference that exists between the values included within the same variable. The dispersion also includes the average of these values.
- Average: is the value that results from the sum of all the values included in the same variable and the subsequent division of the result by the number of data included in the sum. It is defined as the central tendency of a variable.
- Bias or kurtosis: it is the measurement that indicates how steep a curve is. It is the value that indicates the number of elements that are closest to the average. There are three different types of bias (Leptokurtic, Mesocurtic, and Platicúrtic), each of them indicating how high the data concentration is around the mean.
- Graphics: are the graphic representation of the data obtained from the analysis. Usually, different types of statistical graphs are used, including bar, circular, linear, polygonal, among others, - Asymmetry: it is the value that shows how the values of the same variable are distributed in relation to the average. It can be negative, symmetric or positive (Formulas, 2017).
Inferential statistics
It is the analysis method used to make inferences about a population, taking into account the data thrown by descriptive statistics on a segment of the same sample. This segment must be chosen under rigorous criteria.
Inferential statistics makes use of special tools that allow you to make global statements about the population, from the observation of a sample.
The calculations carried out by this type of statistic are arithmetic and always allow for a margin of error, which is not the case with descriptive statistics, which is responsible for analyzing the entire population.
For this reason, inferential statistics requires the use of probability models that allow you to infer conclusions about a large population based solely on what a part of it tells you (Vaivasuata, 2015).
According to descriptive statistics, it is possible to obtain data from a general population from the analysis of a sample made up of randomly selected individuals.
Categories
Inferential statistics can be classified into two large categories described below:
- Hypothesis tests: as its name indicates, it consists of testing what was concluded about a population from the data obtained by the sample.
- Confidence intervals: these are the ranges of values indicated within the sample of a population to identify a relevant and unknown characteristic (Minitab Inc., 2017). Due to their random nature, they are what allow us to recognize a margin of error within any inferential statistical analysis.
Differences between descriptive and inferential statistics
The main difference between descriptive and inferential statistics is that the former seeks to order, summarize and classify the data derived from the analysis of variables.
For its part, inferential statistics, carry out deductions based on previously obtained data.
On the other hand, inferential statistics depends on the work of descriptive statistics to carry out its inferences.
In this way, descriptive statistics constitutes the basis on which inferential statistics will subsequently carry out its work.
It is also important to note that descriptive statistics are used to analyze both populations (large groups) and samples (subsets of populations).
While inferential statistics is responsible for studying samples from which it seeks to reach conclusions about the general population.
Another difference between these two types of statistics is that descriptive statistics only focus on the description of the data obtained, without assuming that they have any relevant property.
This does not go beyond what the data obtained can indicate. For its part, inferential statistics believes that all data derived from any statistical analysis depend on external and random phenomena that can alter its value.
References
- Formulas, U. (2017). Universe Formulas. Obtained from ASYMMETRY: universoformulas.com
- Fortun, M. (June 7, 2012). Statistics Obtained from DESCRIPTIVE AND INFERENTIAL STATISTICS: materiaestadistica.blogspot.com.co
- Minitab Inc. (2017). Retrieved from What is a confidence interval ?: support.minitab.com
- Santillán, A. (September 13, 2016). Evidence. Obtained from Descriptive and inferential statistics: general concepts: ebevidencia.com
- (December 6, 2015). Math. Obtained from Difference between Descriptive Statistics and Inferential Statistics: differenceentre.info