Problem of Index

An Index is often used to measure something that is not easy be quantified or measured. Especially in the field of social science, indices have been widely used for measuring some social phenomena. In some research papers, researchers use indices to measure a country’s government size, degree of democracy and many other subjective characteristics. After these researchers get these indicesm, most of them use the indices to form quantitative analysis models to measure the correlations between the “key indices” and the topic they are studying.

The problem of such indices is that it does not form a complete quantity result. Usually papers using indices may conclude 1 unit change in the index would have a certain impact. However, because an index contains too much information, we have no idea about if all the infomation with the same weights in the index would have the same effect. Moreover, it does not provide useful guidance in practice, as if is hard to make 1 unit change in the index in the real world, as too many factors that would interact with each other and influence the index.

In addition, using different indices can get different results, and researchers tend to use the indices that deliver the most significant results. Such results could be biased and provide inaccurate guidance. Moreover, there is not a standard that measures the accuracy of the indices; therefore, it is very subjective to argue whether an index is good or bad.

Overall, I do not think indices are very useful in research to deliver reliable results.