CSDSNewsletterContact
Programme for Comparative DemocracyCentre for the Study of Developing Societies, Delhi

Assembly Election Study 2012

Lokniti in collaboration with CNN IBN and The Week conducted a Post Poll Survey in the states of Manipur, Punjab, Uttarakhand and Uttar Pradesh. 

A total of 1200 persons in Manipur, 3250 in Punjab, 1680 in Uttarakhand and 6860 in Uttar Pradesh randomly selected from the latest electoral rolls were interviewed few days after the voting in sampled Polling Stations spread across randomly selected Assembly constituencies in the states. The Assembly Constituencies were selected using the probability proportionate to size method. 

Within each selected assembly constituency, 4 polling booths were randomly selected using the Systematic Random Sampling technique. The respondents were selected from the updated electoral rolls of the selected polling stations. Keeping in mind, the probability of non completion of interviews amongst all the selected respondents, for several reasons (Non contact, refusal etc) we adopted the technique of over sampling of respondents. No substitution was allowed for those respondents who could not be contacted or could not be interviewed. 

The interviews were conducted by specially trained field investigators. Before sending investigators to the field for data collection, they were imparted one/two days intensive training in survey research and data collection. The respondents were interviewed in the face-to-face interview situation using a structured interview schedule translated in to the respective state languages. Respondents were mostly interviewed at their home with few exceptions of respondents being interviewed at their work place. The voting question was asked using a dummy ballot paper and dummy ballot box to ensure top most secrecy about this voting preference. 

These are details of the spread of the sample and sample profile

The spread of the sample
 ManipurPunjabUttarakhandUttar Pradesh
No of Assembly Constituencies where survey was undertaken 20 45 30 100
No of Locations where interviews were conducted 80 178 116 381
Total no of voters sampled for interview 2560 5696 3712 12192
Total no of completed interviews 1200 3250 1680 6860
The sample profile: How Representative is the Sample?
 UrbanWomenDalitAdivasiUrban
 CensusSurveyCensusSurveyCensusSurveyCensusSurveyCensusSurvey
Manipur 26.6 31 49.4 50 - - 34.0 31 8.8 3
Punjab 33.9 31 46.7 45 28.9 33 - - Sikh
60.0
Sikh
67
Uttarakhand 25.7 22 49 45 17.9 16 - - Muslim
11.4
Muslim
12
Uttar Pradesh 20.8 16 47.3 40.2 21.2 22 - - 18.1 22

Census figures are from 2001 Census for every state

  • The Assembly constituencies in all the states were selected using the probability proportionate to size method.
  • Polling station (locations) within each selected assembly constituency were selected using the Systematic Random Method
  • Voters were selected using the systematic random method, from the updated electoral rolls of the selected polling booths.
  • The voters were interviewed few days after the voting, in the face to face interview situation mostly at their home or at their work place using a structured Interview Schedule translated in local language.
  • The field investigators, mostly students were imparted training in survey research before being sent for data collection
The survey was designed and analysed by a team of researchers at Centre for the Study of Developing Societies, Delhi. Sanjay Kumar of the CSDSdirected the survey. The fieldwork was coordinated by S.Mangi Singh (Manipur), Ashutosh Kumar and Jagroop Singh Sekhon (Punjab),Annapurna Nautiyal and Jaya Pande (Uttarakhand) and Anil Kumar Verma, Mirza Asmer Beg and Sudhir Kumar(Uttar Pradesh). The team that designed coordinated and analyzed the survey at CSDS comprised of Banasmita Bora, Dhananjai Kumar Singh, Himanshu Bhattacharya, Jyoti Mishra, K.A.Q.A Hilal, Kanchan Malhotra, Kinjal Sampat, Rupali Warke, Shreyas Sardesai, Sohini Mookherjee, Vibha Attri, , and Yogendra Yadav.

Questionnaire

SPECIAL STATISTICS: 2012 STATE ELECTIONS