What was the methodology used?

The sampling design adopted was multi-stage systematic random sampling.

Updated - May 04, 2021 12:27 am IST

Published - May 04, 2021 12:06 am IST

The findings presented here are from post-poll surveys conducted in Assam, Kerala, Tamil Nadu and West Bengal by the Lokniti programme of CSDS, Delhi.

The surveys were conducted from March 28 to April 21 in Assam, April 7 to April 13 in Kerala, April 7 to April 20 in Tamil Nadu, and March 28 to May 1 in West Bengal.

In Assam, the survey was conducted among 3,490 voters at 140 polling stations spread across 35 Assembly constituencies; in Kerala, among 3,424 voters at 140 polling stations of 35 Assembly constituencies; in Tamil Nadu, among 4,354 voters at 160 polling stations in 40 Assembly constituencies; and in West Bengal, among 4,223 voters at 200 polling stations spread across 50 Assembly constituencies.

Sampling design

The sampling design adopted was multi-stage systematic random sampling. The Assembly constituencies were randomly selected using the probability proportional to size method. Thereafter, four polling stations within each of the sampled constituencies were selected using the systematic random sampling method. Within each polling station, 40 voters were randomly sampled from the electoral roll using the systematic random sampling method. Of these 40, 25 interviews were targeted.

The interviews of electors were conducted face-to-face at their homes after voting had taken place in their area. The questionnaire designed for conducting the interview was a standardised semi-structured one and was translated into the local language/s.

In Assam, the interviews were conducted in Assamese and Bengali, in Kerala in Malayalam, in Tamil Nadu in Tamil, and in West Bengal in Bengali.

The interview duration across all States was about 35 minutes on average. While the interview targets were more or less achieved in Assam, Kerala and Tamil Nadu, in West Bengal, the total number of interviews fell short of the target due to challenges/slowdown in fieldwork on account of a surge in COVID-19 cases during the latter phases.

In order to ensure representativeness and correct for under-representation of key demographics, the achieved raw sample has been weighted by gender, religion, locality, and caste group based on Census 2011 data. The final data sets have also been weighted by the actual vote shares secured by the major parties and fronts that contested the elections in each State.

All analysis here has been presented on the weighted data sets.

Safety of investigators

In order to ensure the safety of the field investigators amidst the COVID-19 pandemic and also of those around them, field investigators were provided with masks and sanitisers for the fieldwork and it was ensured that the interviews were conducted following proper physical distancing and mask rules. Before going to the field, field investigators were made to sign an undertaking that they are not showing any of the COVID-19 symptoms and that they would take all the necessary precautions in the field.

In Assam, the survey was coordinated and supervised by Dhruba Pratim Sharma (Gauhati University) and Nurul Hassan (Kampur College, Nagaon).

In Kerala, it was conducted by Sajad Ibrahim, Kiran Raj and Abhishek P.S. (University of Kerala).

In Tamil Nadu, it was conducted by P. Ramajayam and D. Kirubanithi (Bharathidasan University, Tiruchirappalli), Gladston Xavier and Paul K. Nathan (Loyola College, Chennai), Benisha Catherin (Programme manager, Glo Foundation) and Mercy Pious (Independent development consultant).

In West Bengal, it was conducted by Suprio Basu (University of Kalyani) and Jyotiprasad Chatterjee (Barrackpore Rastraguru Surendranath College).

Sanjay Kumar of Lokniti-CSDS, Delhi directed the survey.

0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in


Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.