blr_p
Oracle
http://www.gfilesindia.com/replace-paperless-evms-with-vvpat-evms/
He's asking that the sample size for VVpat be increased from the present one per constituency to 25%
He also wants the failure rate of vvpats to be brought down from 25% to 2-4% which is the accepted rate of EVM failure currently
The man in question
He's asking that the sample size for VVpat be increased from the present one per constituency to 25%
the sample size of ‘one polling station per Assembly Constituency’ is statistically unsound not only in terms of its size but also because the various polling stations in the constituency are not similar and the sample drawn may not be truly representative of the constituency as a whole. There is an imperative need for stratified sampling with a random sample of one or more polling stations drawn from each of the following ‘strata’: urban; semi-urban; rural; those in remote hilly/desert/forest areas; those with very heavy voter turnout (> 80%); those with moderate voter turnout (50% to 80%); those with low voter turnout (<50%); those about which a large number of complaints were received, and so on.
We, therefore, suggest that VVPAT slips must be simultaneously counted for a sample size of at least 25% of the polling stations in an Assembly Constituency with the samples drawn randomly from the different strata and verified with the electronic count. If any variation is found then the entire VVPAT slips in the constituency should be counted and tallied with the electronic count before declaring the result. We believe this is a direction that can be issued by your office for the upcoming 2019 general elections, in order to a greater standard of fairness and transparency in the electoral process.
He also wants the failure rate of vvpats to be brought down from 25% to 2-4% which is the accepted rate of EVM failure currently
The man in question