THE CONTEXT: The importance of statistical thinking in the context of the Supreme Court of India’s decision on the sample size for Voter Verifiable Paper Audit Trail (VVPAT)–based audit of Electronic Voting Machines (EVMs) cannot be ignored. It emphasizes that truths in science, mathematics, and statistics cannot be established through legislative or judicial fiats, highlighting the need for statistically significant sample sizes in the audit of EVMs to ensure election integrity.
ISSUES:
- Inadequacy of Sample Size: The Supreme Court’s decision on the statistically inadequate sample size for VVPAT-based audits of EVMs is questionable. The prescribed sample size does not conform to the fundamental principles of statistical sampling theory, potentially failing to detect defective EVMs.
- Lack of Transparency: The Election Commission of India (ECI) and the Supreme Court lack clarity and transparency regarding the definition of the ‘population’ of EVMs for sampling and the subsequent steps if a defective EVM is found.
- Risk of EVM Malfunction or Manipulation: Despite claims of no mismatches, EVMs are susceptible to malfunction or manipulation, and the current audit system may not adequately address or reveal these issues due to the small sample size.
- Judicial and Administrative Oversight: The judiciary and administrative bodies are being criticized for not demanding more rigorous statistical methods and transparency in the auditing process of EVMs.
- Misunderstanding of Statistical Principles: Political figures and parties are criticized for demanding arbitrary percentage samples for EVM audits, which reflects a misunderstanding of how statistical accuracy is achieved.
- Need for a Statistically Sound Audit System: A statistically sound VVPAT-based audit system that can detect mismatches with high accuracy is quintessential. Results should be declared based on EVM counts only where there is a perfect match. Otherwise, manual counting of VVPAT slips should be mandated.
THE WAY FORWARD:
- Define the Population: Clearly define the ‘population’ of EVMs for each Assembly and Parliamentary Constituency. This clarity is essential for determining the appropriate sample size and subsequent steps in case a defective EVM is detected.
- Determine Statistically Significant Sample Size: Use statistical sampling theory to determine a sample size that can detect defective EVMs with high confidence (99% or 99.9%). This involves calculating the sample size based on the population size, the acceptable risk level (type I and type II errors), and the expected rate of defective EVMs.
- Random Sampling: Ensure that the sample of EVMs for VVPAT verification is selected randomly from the defined population. This randomness is crucial for the integrity and reliability of the audit process.
- Immediate Verification: Conduct the matching exercise between EVM counts and VVPAT slips at the beginning rather than at the end of the counting day. This allows for immediate action in case of discrepancies.
- Action in Case of Mismatch: If a mismatch between the EVM count and the VVPAT count is detected in the sample, manual counting of VVPAT slips should be conducted for all remaining EVMs of the ‘population’. The results should then be declared based on the VVPAT count.
- Transparency and Public Trust: Transparently disclose the methodology used for sampling, the results of the VVPAT verification, and the actions taken in case of mismatches. This transparency is vital for building and maintaining public trust in the electoral process.
- Continuous Review and Improvement: Regularly review and update the sampling methodology and audit process based on technological advancements, electoral landscape changes, and stakeholder feedback. This ensures that the system remains robust and effective in detecting and addressing any issues with EVMs.
THE CONCLUSION:
The Supreme Court’s decision not to demand transparency from the Election Commission of India regarding the definition of ‘population’ for sample size is criticized, and the steps to be taken in case of a mismatch are also criticized. It argues for implementing a statistically sound VVPAT-based EVM audit system that can detect mismatches with high accuracy, advocating for a ‘management by exception’ approach to ensure the integrity of election results.
UPSC PAST YEAR QUESTION:
Q. In the light of recent controversy regarding the use of Electronic Voting Machines (EVM), what are the challenges before the Election Commission of India to ensure the trustworthiness of elections in India? 2018
MAINS PRACTICE QUESTION:
Q. Discuss the significance of adopting a statistically sound VVPAT-based system for EVM audit in ensuring the integrity of electoral processes. Evaluate the implications of the Supreme Court’s verdict on the sample size for the VVPAT audit in the context of the Association for Democratic Reforms vs Election Commission of India and Another (2024). Suggest measures to enhance the reliability of the electoral audit process.
Spread the Word