CHENNAI: The Election Commission of India (ECI), in its communication dated October 27 to kickstart the Special Intensive Revision (SIR) of electoral rolls in nine states and three union territories, said electors whose enumeration forms have not been returned could be dropped from the draft rolls for one of four reasons — absent, shifted, deceased or duplicate.
However, when the draft rolls were published in these places, including Tamil Nadu, the commission did not provide disaggregated data on absent and shifted. Instead, the two were clubbed. Of the 97.4 lakh names deleted in Tamil Nadu, 26.9 lakh were marked as deceased, 3.98 as duplicates and another 66.4 lakh as ‘shifted/absent’.
An analysis of reason-wise disaggregated data of all 97.4 lakh deleted records through software tools by TNIE revealed that 52.7 lakh of the 66.4 lakh were marked as shifted and the remaining 13.7 lakh as absent.
Importantly, constituency-wise and booth-wise trends and enquiries with officials showed wide variations and lack of clarity, which led to the two being used interchangeably and arbitrarily, particularly in urban areas. Moreover, TNIE found ‘shifted’ being used as a blanket reason for deletion of hundreds of voters in many booths.
According to senior officials, shifted and absent are to be used in distinct scenarios. Shifted is when a booth-level officer (BLO) has reliably learnt that the person has permanently moved out, while absent (or not residing as mentioned in deletion lists) meant that the person remained an ‘ordinary resident’ of the place, but could not be reached/traced during the enumeration phase.
However, data showed inconsistency in understanding and use of the two. Extreme examples at the constituency level are Harbour and Anna Nagar in Chennai and Vellore constituency. Harbour, which dropped 69,824 electors in the draft rolls, had ‘zero’ deletions due to absent, but had 60,258 (86.3%) as shifted.
Anna Nagar, which dropped 1,18,287 (42.2% of its electorate) voters in the roll, had just one person marked as absent, but 1,08,828 (92%) marked as shifted. At the other end is Vellore, where 29,043 (53.2%) of the 54,565 deleted were marked as absent.
Enquiries with officials from these constituencies revealed differences in their understanding. An official from Chennai said they asked BLOs to mark untraceable persons or addresses as shifted. With house numbers having changed in many places, many BLOs said they marked numbers they could not match or trace as shifted.
The pattern indicated that Vellore perhaps displayed better fairness by marking untraceable or unreachable persons as ‘absent’. When asked if they are able to get more enrolment forms from persons marked as absent, since in theory they continued to be ordinary resident of the constituency, an election official from Vellore said, “We have received around 6,000 forms, but a majority of them are from new voters who were not present in earlier rolls.”
TNIE found at least six booths in Chennai and Tiruchy districts, where 322 to 584 deletions were done with everyone marked as shifted — not a single, deceased, duplicate or absent. The official in Tiruchy said it was a residential area where several government employees had shifted out and the marking was done with verification by senior officials.
However, a BLO from Anna Nagar, which saw high deletions, said she had been receiving several calls from voters who have been marked as shifted, indicating possible deletions with blanket application of ‘shifted’ as reason.
The statistical measurement of Pearson correlation within urban, semi-urban and rural constituencies showed the strongest evidence of interchangeable usage of the two reasons due to varying instructions or understanding.
The measurement, which shows whether one variable goes up when the other goes down, showed a strong negative correlation of -0.73 in urban constituencies when such a correlation should not exist in the real scenario. (A value beyond -0.5 indicates a strong negative correlation). The value is -0.36 in rural areas, indicating better application of the reasons there.