B.Com 2nd and 3rd Semester (CBCS Pattern)
Business Statistics Notes
Classification of Data - Univariate, Bivariate and Multi - Variate Data
Classification of Data
The process of organizing data into groups or classes according to their general characteristics is technically known as segregation. Separation to integrate related facts into classes. It is the first step in the construction of a table.
In Secrist's words, "segregation is the process of arranging data in order of groups according to their common characteristics or separating different but related parts."
Essentials of classification
a) The division must be complete so that the entire distribution unit finds a place in one group or another.
b) The planning must be in line with the objectives of the investigation.
c) All the elements that make up a team must be the same.
d) The layout should be expanded so that new facts and figures can be easily adjusted.
e) Planning should be sustainable. If not and converted to all queries then the data will not be worth asking.
f) Data must not be transferred. Each data item must be available in one class.
Naturally, data can be categorized as:
1. Fixed Data: Fixed data is data where there is only one variation. It is a simple type of data and only deals with price change. It does not address the cause-effect relationship between two or more variables. The main purpose of static data is to find an existing pattern in a given data. Examples of consistent data are 10 years of factory workers, marks obtained by 50 students in mathematics etc. The pattern in the static data can be analyzed with the help of various tools for measuring the inclination of medium and scattering. Also drawings and graphs can be used to analyze static data.
2. Bivariate Data: Bivariate data is data where there are two different variants. It deals with the cause and effect relationship between two variables and data analysis is performed to determine the relationship between the two variables. Examples of Bivariate data age and staff weight, product demand and supply, student marks in two different subjects, temperature and sales of wool products etc.
3. Multivariate Data: Multivariate data is data where there are three or more variables. An example of multivariate data is that a developer wants to compare the sales data of his product in five different areas. Another common tool used to analyze multivariate data is correlation and regression analysis, multivariate analysis of variance etc.
4. Time series data: Time series data is a study of the behavior of one variant at different times. This data is being analyzed to determine the trend. An example of time series data is product X sales over a period of 10 years.
5. Cross-sectional data: Cross-sectional data is a study of the behavior of more than one variant at a different time point. An example of the data for different components is the sale of product X over a period of 10 years in four different regions.