Note4Students: 

Prelims: EVM; VVPAT; 

Mains: Landmark Judgements by SC; Judicial Interventions;

Mentor comments: H.G. Wells once said “Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write”. Just as water cannot be legislated to flow uphill, the establishment of statistical principles cannot be arbitrarily defined by authorities. The Supreme Court of India has recently dismissed a batch of petitions requesting 100% verification of votes cast using Electronic Voting Machines (EVMs) with Voter Verifiable Paper Audit Trail (VVPAT) slips. The top court recommended that the Election Commission of India (EC) needs to consider incorporating bar codes for each political party on the paper slips. Meanwhile, PM Modi has called the Supreme Court’s verdict a tight slap on the face of the Opposition raising doubts about EVMs and should now apologize to the nation. So, will the Supreme Court verdict end the EVM trust debate?

Let’s learn

—

Why in the News?

The SC of India cannot arbitrarily mandate that a uniform sample size of “5 EVMs per Assembly Constituency” is good enough for VVPAT-based audit of EVMs for all Assembly Constituencies across the country. 

Context:

• Ensuring the accuracy of votes cast is crucial – while VVPAT verification confirms votes as cast, discrepancies may arise in the counting process. 
• Hence, to mitigate risks of EVM malfunction, it is essential to conduct manual counts of VVPAT slips against EVM tallies for a statistically significant random sample of machines, aligning with fundamental principles of statistical sampling theory to ensure the integrity of the electoral process.

A typical case

• Lot of Acceptance Sampling: The VVPAT-based audit of EVMs follows lot acceptance sampling, a quality control method where a sample is inspected for defects.
  • If defects exceed a set limit, the entire lot is rejected. Defective EVMs are those with discrepancies between EVM and VVPAT counts. 
• Lack of Clarity: The Supreme Court did not specify the ‘population’ of EVMs to which its sample size relates. Further, the SC and ECI also lack clarity on handling defective EVMs in samples.
  • If defects are found, all EVMs from that population should be manually counted. Defining the population as Assembly or Parliamentary Constituencies impacts the detection probability of defective EVMs.

What Election Commission of India (ECI) have claimed?

• The ECI has claimed that there have been no instances of mismatch between the EVM count and the VVPAT count over the years. 
• However, this claim is disputed, and the few mismatches that do occur could be due to multiple reasons like the well-functioning of the EVMs, an inadequate sample size that fails to detect defective EVMs effectively, or a combination of both factors. 

Reasons behind the ECI’s claim

• There are major three reasons:
  • The ECI and the Supreme Court’s endorsement of the current sample size and audit protocol is criticized as flawed. 
  • The Court’s judgment in the Association for Democratic Reforms vs Election Commission of India and Another (2024) is disappointing because it did not compel the ECI to make public how it has defined the ‘population’ to which its sample size relates and its ‘next steps’ in the event of a mismatch.
  • The Court also did not clarify these points on its own after seeking expert opinion.
• Despite rejecting extreme demands like a return to paper ballots or 100% VVPAT verification, the Supreme Court’s prescribed sample size in 2019 was criticized as arbitrary and incorrect, highlighting the need for more statistically sound approaches in EVM audits.

What needs to be done?

• Need for Sound Audit System: We do not need to know the various ways an EVM may fail or be manipulated. What is required is a statistically sound VVPAT-based EVM audit system that can detect mismatches with 99% or 99.9% accuracy.
  • By implementing a robust, the integrity of the electoral process can be ensured without resorting to extreme measures like a return to paper ballots or 100% VVPAT verification.
• Need to achieve Golden Mean: The VVPAT-EVM count matching should be done at the beginning of the counting day, not at the end. This statistical sampling-based, ‘management by exception‘ approach represents the golden mean as follows:
  • Where there is a perfect match, the results should be declared based on the EVM count.
  • Only where there is a mismatch, manual counting of VVPAT slips for the entire ‘population’ of EVMs should be done, and the results declared based on the VVPAT count.