Page 42 - DCOM203_DMGT204_QUANTITATIVE_TECHNIQUES_I

Unit 2: Classification of Data

2.5 Summary Notes

Classification of data on the basis of one, two or more factors is termed as a one-way, two-
way or multi-way classification, respectively.

Classified data, when arranged in some logical order such as, according to size or according
to time of occurrence or according to some other criterion, is known as statistical series.
A statistical series, in which data are arranged according to magnitude of one or more
characteristics, is known as a frequency distribution.
Data classified according to the magnitude of only one characteristic is known as uni-
variate frequency distribution.
Data classified, simultaneously, according to the magnitude of two or more characteristics
are known as bivariate or multivariate frequency distributions respectively.
When a characteristics is an attribute, the data can be classified into two or more classes
according to this attribute, known as dichotomous or manifold classification respectively.

When the data are simultaneously classified according to two or more attributes, the
classifications are two-way or multi-way respectively.
It is also possible to have a two-way or multi-way classification in which one or more
characteristics are variables while others are attributes.
The process of classification is facilitated by writing the classified data in tabular form.
Using tables, it is possible to write huge mass of data in a concise form. Further, it helps to
highlight essential features of the data and make it fit for further analysis.

2.6 Keywords

Bivariate frequency distributions: Data classified, simultaneously, according to the magnitude
of two characteristics are known as bivariate frequency distributions

Classification: Classification is the process of arranging things (either actually or notionally) in
the groups or classes according to the unity of attributes that may subsist amongst a diversity of
individuals.
Dichotomous classification: When a characteristic is an attribute, the data can be classified into
two classes according to this attribute, known as dichotomous classification.
Frequency distribution: A statistical series, in which data are arranged according to magnitude
of one or more characteristics, is known as a frequency distribution.

Manifold classification: When a characteristics is an attribute, the data can be classified into two
or more classes according to this attribute, known as dichotomous or manifold classification
respectively.

Multivariate frequency distributions: Data classified, simultaneously, according to the magnitude
of more than two characteristics are known as multivariate frequency distributions.
Statistical series: Classified data, when arranged in some logical order such as, according to size
or according to time of occurrence or according to some other criterion, is known as statistical
series.
Uni-variate frequency distribution: Data classified according to the magnitude of only one
characteristic is known as uni-variate frequency distribution.

LOVELY PROFESSIONAL UNIVERSITY 37