What is classification of data? What should be its characteristics?
What is classification of data? What should be its characteristics?
Classification is an ordered set of related categories used to group data according to its similarities. It consists of codes and descriptors and allows survey responses to be put into meaningful categories in order to produce useful data. To be meaningful, classification should have the following characteristics.
(i) It should be unambiguous - Classification aims at removing ambiguity. It is a must that all classes should be defined in such a way that there is no confusion and each item must fit to at least and at most one class.
(ii) The classes must not overlap - None of the items should be eligible to be a part of more than one class.
(iii) It should be stable - Without stability, classified data will not be fit for comparison.
(iv) Classification should be according to the purpose of enquiry. For example, if I need to classify the students into two groups for bus arrangement, it will be better to use geographical classification. If the purpose is judging their academic performance the quantitative classification is more suitable. If the purpose is judging their value system then qualitative classification is recommended.
(v) It should be mathematically accurate. The test of mathematical accuracy is confirmation of total items in the series with total items in the universe.
(vi) It should be flexible. It should be possible to adjust the series to new situations and circumstances. With a change in time, some figures may become obsolete and other may become more relevant.
Classification is an ordered set of related categories used to group data according to its similarities. It consists of codes and descriptors and allows survey responses to be put into meaningful categories in order to produce useful data. To be meaningful, classification should have the following characteristics.
(i) It should be unambiguous - Classification aims at removing ambiguity. It is a must that all classes should be defined in such a way that there is no confusion and each item must fit to at least and at most one class.
(ii) The classes must not overlap - None of the items should be eligible to be a part of more than one class.
(iii) It should be stable - Without stability, classified data will not be fit for comparison.
(iv) Classification should be according to the purpose of enquiry. For example, if I need to classify the students into two groups for bus arrangement, it will be better to use geographical classification. If the purpose is judging their academic performance the quantitative classification is more suitable. If the purpose is judging their value system then qualitative classification is recommended.
(v) It should be mathematically accurate. The test of mathematical accuracy is confirmation of total items in the series with total items in the universe.
(vi) It should be flexible. It should be possible to adjust the series to new situations and circumstances. With a change in time, some figures may become obsolete and other may become more relevant.
