Electoral Politics in the Indian States : Three Disadvantaged Sectors/ by Jagdish N. Bhagwati... (et al )

The M.I.T. Indian Election Data Project was began in early 1968 initially under a pilot grant from the Center for International Studies, followed by a major grant from the National Science Foundation, with the objective of undertaking a series of computer-based studies of elections in India since 1952 in the more than 3,000 state amembly constituencies The early ambitious hope was that these studies could bring to the analysis of Indian elections some of the methodologies and sophisticated statistical tools that have been developed for the study of elections in the United States and other developed countries, test with the Indian electoral data some of the general propositions that have emerged in recent studies of political development, improve our knowledge of the world's largest democratic state, and enhance our capacity to predict future electoral change in India.

The technical dimensions of this project proved to be so formidable that at times we despaired of producing any studies! The data had to be computerized, "cleaned," and checked for inconsistencies. They then had to be arranged in files, concepts had to be converted into measures, and innumerable problems of how to compare constituencies with one another and with themselves over time had to be resolved. At an early stage in the project we considered matching selected census data to constituencies in order to relate some electoral variables such as turnout, competitiveness, and party performance to socio-economic variables but decided not to do so since a number of such studies were under way elsewhere-particularly the work of W.H. Morris-Jones and Biplab Das Gupta at the Institute for Commonwealth Studies in London, Rajni Kothari at the Centre for the Study of Developing Societies in Delhi, Paul Brass at the University of Washington, Harry Blair at Bucknell, and Donald Zagoria at Columbia University.