Lok Sabha Election 2019: News18-IPSOS exit poll used four stages of multi-sampling as part of its methodology

Polling will be held on Sunday in the seventh and final phase of the Lok Sabha elections 2019. Voting will take place in all 13 seats in Punjab and an equal number of seats in Uttar Pradesh, nine in West Bengal, eight seats each in Bihar and Madhya Pradesh, four in Himachal Pradesh, three in Jharkhand and the lone seat Chandigarh. [caption id=“attachment_6533781” align=“alignleft” width=“380”] Representational image. PTI[/caption] After the final phase is over, News18 — with research partner IPSOS — will present its exit poll to its viewers. To predict the number of seats likely to be won by each party or alliance, a total of 199 parliamentary constituencies were selected out of 543. Within the parliamentary constituencies, 796 Assembly constituencies were selected, and 4,776 individual polling station areas were selected for conducting interviews. In each polling station, around 25 voters were randomly selected depending on the terrain and voter turnouts in that particular polling station. A total of 1,21,542 voters were interviewed from 199 parliamentary constituencies. Multi-sampling process was adopted for the selection of parliamentary constituencies, Assembly constituencies and polling stations. This process was divided into four stages. In the first stage, the selection of parliamentary constituencies was done on the basis of past election results. In the second stage, four to six Assembly constituencies were selected in each parliamentary constituency using systematic random sampling (SRS) process. Click here to follow EXIT Poll 2019 Results In the third stage, in each Assembly constituency, six polling stations were selected. And in the fourth stage, voters were randomly selected outside polling stations. The survey was conducted on election day in all seven phases of polling. The interviewing process started when the polls opened and continued throughout the day until polls closed to ensure better representation of all voters. Each polling booth was covered for one hour by three interviewers. They then moved to the next polling booth. All interviews were conducted using computer-assisted personal interviewing (CAPI) devices. The questionnaire was translated into the vernacular language which is mainly spoken in the respondents’ state. Data was weighted to reflect the constituency profile on age, gender, caste, religion and voter turnout.