Association of attributes

In social sciences, we come across certain phenomena which are incapable of quantitative measurement.Blindness, deafness, religion; juvenile delinquency, marital status etc are some phenomena which are not measurable. Such characteristics are called attributes. In these cases, one can make only counting of individuals who possess or do not possess these attributes. In other words what can do is to state so many individuals are blind or so many non-blind. While dealing with one attribute the classification of data is done on the basis of presence or absence of the attribute. It is also absolutely essential that a clear-cut definition of the attribute under study is made because only such a definition paves way for the counting of the individuals possessing or not possessing the attribute. Two attributes are said to be associated only if they appear together in a great number of cades than is to be expected if they are independent. On the other hand, if the number of observed cases is less than the expected, under assumption of independence, attributes are associated. In order to ascertain whether the attributes are associated or not the following methods can be used.

1. Comparison of observed and expected frequencies.
2. Proportion method
3. Yule's coefficient of Association
4. Coefficient of colligation
5. Coefficient of contingency