30 seats, 2,079 respondents: How Lokniti survey was done

Field investigators asked the respondents, face to face, a detailed set of questions which could take up to 20-25 minutes. The achieved sample largely matches the actual demographic profile of the population in Bihar.

This analysis is based on a 2015 pre-poll survey conducted by Lokniti, Centre for the Study of Developing Societies, Delhi. It was conducted among 2,079 respondents in 120 locations across 30 constituencies. These constituencies were randomly selected using the “probability proportionate to size” method. Thereafter, four polling stations within each of the sampled constituencies were selected using the “systematic random sampling method”.

See chart

The respondents, too, were randomly selected using the same method from the latest rolls of the sampled polling stations. The fieldwork was conducted between September 25 and 30. Field investigators asked the respondents, face to face, a detailed set of questions which could take up to 20-25 minutes. The achieved sample largely matches the actual demographic profile of the population in Bihar. Wherever there was a major under-representation of any important category, we weighted the data file with census figures.

The fieldwork of the study was coordinated by Rakesh Ranjan (Patna College, Patna). The survey was designed by a team of researchers at Lokniti-CSDS. The team included Vibha Attri, Himanshu Bhattacharya, Nitin Mehta, Jyoti Mishra and Dhananjay Kumar Singh.