What Election Commission told Mallikarjun Kharge on revised voter turnouts, how figures are arrived at

What were the initial and final turnout figures in the first two phases of polling?

In an official release issued at 7.55 pm on April 19, the ECI said turnout in the constituencies where votes were cast in the first phase was “over 60% at 7pm”. The release said the final figure was likely to be higher after reports came in from polling stations where voting continued beyond the scheduled cut-off of 6 pm to allow voters who had already entered the stations to vote.

A day later, most reports, including those in The Indian Express, said turnout was 65.5%, attributing it to senior officials in the ECI.

On April 26, when votes were cast in the second phase, the ECI press release, issued at 9 pm, put the turnout at 60.96% at 7 pm. News organisations subsequently reported a more-or-less final figure of 66.71%, based on unofficial confirmation from senior ECI officers.

On April 30, the ECI released the final figures for both phases through an official release, which put the phase 1 turnout at 66.14% and the phase 2 turnout at 66.71%.

After voting in the third phase on May 7, the ECI first released an estimated turnout of 61.45% at 8 pm, and then revised it some time before midnight to 64.4% for 11.40 pm. The day after, this turnout figure was updated to 67.33%. In its first press release on May 7 evening, the ECI said it would release the final turnout data by May 11.

Why is there a difference between initial and final voter turnout numbers?

According to the ECI, the initial number is a preliminary estimate, which changes. ECI officials say the higher final figure does not mean that votes were cast even after polls closed — it could be simply due to the fact that the Returning Officers (ROs) sent in the updated data late.

According to the ECI officials, even the data released on April 30 is not final — that figure will be known only after postal ballots are counted on the day of counting (June 4) and added to the turnout.

On both April 19 and April 26, state- and constituency-wise turnouts were available in real time on the ECI’s Voter Turnout mobile application. This app — which was launched by the ECI after the second phase of polling in 2019 — carries this disclaimer: “This is an approximate trend, as data from some Polling Stations (PS) takes time and this trend does not include data of postal ballot voting. Final data for each PS is shared in Form 17C with all Polling Agents.”

Form 17C is made available to all candidates through their polling agents at the close of polling. Form 17C includes the total numbers of electors and voters at the seat. This information is not available on the Voter Turnout app. (More on Form 17C below)

Were the official turnout figures — released 11 days and 4 days after phases 1 and 2 of the voting respectively — unusually delayed?

No, says the ECI. In its response to Kharge, the ECI provided details of the release of official turnout data for the 2019 Lok Sabha election and state Assembly elections over the past few years. This data, it said, “would testify there is no unusual delay than what is required to collect data from large number of polling stations”. The data provided by the ECI showed a 1-3 percentage point increase in the turnouts in the 5-7 days it took to release the phase-wise turnouts in 2019.

ECI officials say it takes time to verify and collate the data, particularly from remote polling stations. In some cases, polling parties have to make it back from these remote locations to the district headquarters, where they have a stable Internet connection to transmit the data.

In its letter to Kharge, the ECI said that after the polling parties arrive, documents are scrutinised by the Returning Officers in the presence of candidates and observers. The RO then completes data entry, and publishes the data on the Voter Turnout app.

“This process takes P+1 (polling day plus 1) or P+2 (polling day plus 2) days depending on constituencies and arrival schedule of polling parties due to geographical and weather conditions. This data may further get updated for any constituency where repoll is being conducted, on conclusion of repoll and arrival of parties which is normally on P+4 days,” it said.

And what is Form 17C? To whom is it issued, and when?

As per the Conduct of Elections Rules, 1961, there are two forms that have data on the number of electors and voters — Forms 17A and 17C.

The former is a register of voters, in which polling officials record the details of every voter who comes into the booth and signs the register. The latter is the account of the votes recorded.

Form 17C is issued to the polling agents of candidates at the close of polling.

It has the identification numbers of the EVMs used in the polling station; the total number of electors assigned to the polling station; the total number of voters as entered in the register for voters (Form 17A); the number of voters who decided not to record their votes after signing the register; the number of voters who were not allowed to vote; and the total number of test votes to be deducted, and total number of votes recorded per EVM.

Part-II of the same form carries the results of the counting, which are entered on counting day.

The data in Form 17C are considered final, and lie at the heart of any Election Petition that may be filed challenging the result. In a statement issued on April 30, the ECI said the “actual data of Form 17C shall prevail which is already shared with candidates”.

Apart from being the record that is to be used in a possible legal challenge, the data in Form 17C are also used by candidates to verify the results on counting day. The number of electors and those who cast their votes as recorded in Form 17C can be matched with the EVM count.

Why is there a demand to make the Form 17C data public?

Activists have sought to make a case for full transparency. Also, they say, even if Form 17C is made available to the candidates’ agents, given the large numbers of polling stations at each constituency, difficulties of coordination, and shortage of resources, many candidates find it impossible to collate all the figures for the constituency as a whole.

The ECI has said that it is not “legally bound to publish any voter turnout data at aggregate level”. It has sought to allay concerns by reiterating that the provision of Form 17C is a statutory requirement, and no candidate or party has alleged that they have not received the form at the close of polling.

Also, on the day of counting, the counting supervisor must certify in Part-II of the form that there is no discrepancy in the votes counted and the votes recorded in Part-I on polling day. This is countersigned by all candidates or their representatives, and discrepancies, if any, can be challenged, the ECI says.