Poll methodology
THE findings of the Indian Express-CNN IBN-CSDS Poll reported today are based on the Mid-Term Poll Verdict conducted in 18 states of the country between 1-4 September 2007. During the survey, 18,750 respondents were interviewed by 489 field investigators and Supervisors spread across 940 polling stations from a sample of 217 Lok Sabha constituencies.
A three-stage stratified random sampling technique (Probability Proportionate to Size (PPS) method) was used to select the respondents for this purpose. The respondents who were interviewed during the survey were drawn randomly using the electoral rolls of the sampled polling booths.
In a face-to-face situation the respondents in all the 18 states were interviewed at their place of residence using a structured questionnaire in the local dialect of the state in which he or she resided. The sample comprised 45 per cent women, 79 per cent Hindus, 11 per cent Muslims, 18 per cent Dalits, 7 per cent Adivasis and 73 per cent rural respondents.
The fieldwork for the survey was coordinated and conducted in the states by P. Narasimha Rao (Andhra Pradesh), Sandhya Goswami (Assam), Rakesh Ranjan (Bihar), Baba Mayaram (Chhattisgarh), Kinjal Sampat (Delhi), Priyavdan M Patel (Gujarat), Dr Harish Kumar (Haryana), Harishwar Dayal (Jharkhand), Padmavathi B. S. (Karnataka), Sajad Ibrahim (Kerala), Ram Shankar (Madhya Pradesh), Nitin Birmal (Maharashtra), Surya Narayan Mishra (Orissa), Jagroop Singh Sekhon (Punjab), Sanjay Lodha (Rajasthan), Gundapuneni Koteswara Prasad (Tamil Nadu), A. K. Verma (Uttar Pradesh Central), Sudhir Kumar (Uttar Pradesh East) and Suprio Basu (West Bengal). Sanjay Kumar of Centre for the Study of Developing Societies (CSDS), Delhi, directed this all-India survey. The Central team, which designed, coordinated and analysed the survey, comprised Yogendra Yadav, Praveen Rai, Sanjeer Alam, Dhananjai Joshi, Vikas Gautam, K. A. Q. A Hilal, Himanshu Bhattacharya, Kanchan Malhotra of CSDS and Rajeeva Karandikar of Cranes Software International Limited.