There are no hard and fast rules for making classification of data. Technically the classification of data depends upon the nature, scope and purpose of the study. Nonetheless, an ideal classification possesses some characteristics. Some of the characteristics are given below:
Classification of data must be unambiguous. Various classes should be so defined that there is no room for doubt for confusion.
(ii) Exhaustive and mutually exclusive
Classification should be done in such a manner that each and every item belongs to only one class. This implies that different classes should not overlap.
Classification should conform to the object of enquiry. For example, if an enquiry is conducted to study the economics condition of the workers of Charge Chrome project at Bhadrak it is of no use of classifying them on the basis of their caste or religion.
Date are classified generally on the basis of some criterion. Once such criterion is fixed, it should be retained for other related matters.
A good classification should be flexible. It should have the capacity to accommodate with the new situation. Stability of classification does not mean rigidity of classes. The term stability is used in a relative sense. No classification can be stable for ever. Changes here and there become necessary with charge in time and other changed circumstances. An ideal classification should be such that it can adjust itself to these changed and yet retain its stability.