How this post-poll survey differs from an exit pollhttps://indianexpress.com/article/india/politics/how-this-post-poll-survey-differs-from-an-exit-poll/

How this post-poll survey differs from an exit poll

The survey in Maharashtra, Haryana was conducted among 1,542 respondents spread across 35 constituencies.

BJP supporters celebrate the party’s unprecedented victory in Pune city, at Chhatrapati Shivaji Sports Complex, Balewadi, where the votes were counted Sunday. (Source: Express photo by pavan Khengre)
BJP supporters celebrate the party’s unprecedented victory in Pune city, at Chhatrapati Shivaji Sports Complex, Balewadi, where the votes were counted Sunday. (Source: Express photo by pavan Khengre)

By: Lokniti Team 

This analysis is based on post-poll surveys conducted by Lokniti, Centre for the Study of Developing Societies, in Maharashtra and Haryana. The survey in Maharashtra was conducted among 1,542 respondents spread across 35 assembly constituencies.

In Haryana it was conducted among 1,462 respondents in 25 constituencies. In both states, the constituencies were randomly selected using the “probability proportionate to size” method.

SAVE1

Thereafter, four polling stations within each of the sampled constituencies were selected using the “systematic random sampling” method. Finally, the respondents were also randomly selected using the same method from the latest electoral rolls of the sampled polling stations.

Advertising

A post-poll survey is very different from an exit poll. Trained field investigators approached voters for an interview during the period after the polls were over (after 6 pm on October 15) and before the results were known on October 19. The investigators asked the respondents, in a face-to-face interview, a detailed set of questions which could take up to 20-25 minutes.

The achieved sample matches largely with the actual demographic profile of the population in the two states. The post-poll studies were not used to make any vote or seat projections but are purely academic exercises for explaining the verdict. The data file used for analysis has been weighted by the actual vote share secured by major parties.

The fieldwork of the study in Maharashtra was coordinated by Nitin Birmal (Dr Babasaheb Ambedkar College, Pune), Vivek Ghotale (Savitribai Phule Pune University), Harshad Bhosale (Kirti College, Mumbai) and Rahul Bavge (Govt. College, Nagpur). The fieldwork in Haryana was coordinated by Kushal Pal and Anita Agarwal (Dyal Singh College, Karnal) and Nitin Mehta (Lokniti-CSDS, Delhi).

The surveys were designed and analysed by a team of researchers at Lokniti-CSDS. The team included Anuradha Singh, Dhananjay Kumar Singh, Himanshu Bhattacharya, Jyoti Mishra, Pranav Gupta, Shreyas Sardesai and Vibha Attri. Suhas Palshikar provided his suggestions during the entire exercise. The surveys were directed by Sanjay Kumar.