what is Classification and objectives and Basic principle of classification in stastitic?

Classification:

Introduction:

It is difficult to draw any inferences form the data that have been originally collected. There are chances of making wrong decisions about the nature of the data, because human mind is not so capable of memorizing all the figures. Thus there arises a need to reduce and simplify the raw data (primary data) into such a form that is easily understood. One such form of reducing the data is classification.

Definition:

The process of arranging data into various groups or classes according to some common characteristics is called classification.

Aims or objectives of classification:

The main objectives of classifying the data are:

                    i.                    Since human mind is not so fertile to remember all the figures. 

                      Therefore classification is the only way to reduce the large mass of data.

                  ii.                  Classification facilitate comparison i.e. when data are classified it becomes easy to know how many students source marks between 20-40, 40-60 etc.

                iii.                  Classification simplifies calculation of statistical measure like mean, median, standard deviation etc.

               iv.                     When data are originally collected, there is repetition of values which consumes too much space and time. The classification technique saves time and space because data are presented in a compact form in comparison to lose form.

Basic principle of classification:

While studying the larger set of data, the following points should be taken into Consideration.

                    i.                   The classes into which data are to be distribute should be mutually exclusive i.e. successive classes should not overlap.

                  ii.                  The classification procedure should be exhaustive i.e. classes should completely cover the whole data. For the proper analysis no item should be left classified.

                iii.                  Classification should be clear and simple. Ambiguities and doubtful entries must be removed.

               iv.                     The classification procedure should not be so slab orate  to lead trivial classes nor it should be so crude as to accommodate whole  data in one or two classes.