ZESN endorses new voters' roll

The Zimbabwe Elections Support Network (ZESN) has given the new voters' roll compiled by the Zimbabwe Electoral Commission (ZEC) through the biometric voter registration (BVR) a vote of confidence after an independent audit it carried out. ZESN is a local non-governmental organisation involved in electoral issues. ZESN chairperson Mr Andrew Makoni said this year's voters' roll was an improvement on the 2013 document.

"Overall, ZESN finds that the 2018 voters' roll received on the 18th of June is an improvement over the 2013 preliminary voters' roll," Mr Makoni said in a statement yesterday. ZESN said the audit involved field and computer tests to check on the accuracy, currency and completeness of the voters' roll.

"ZESN's audit of the 2018 voters' roll had two components: field tests that involved comparing registration information collected from voters with what is on the 2018 voters' roll; and computer tests that entailed analysing the 2018 voters' roll for internal consistency and in comparison to the 2013 preliminary voters' roll and census data," said Mr Makoni.

"The 2018 voters' roll was assessed along three dimensions: Accuracy (the degree to which the voters' roll has errors); Currency (the degree to which the voters' roll has been updated); and Completeness (the degree to which the voters' roll contains all eligible voters)." Mr Makoni said the audit did not identify any anomalies that affected a large percentage of the registrants.

"In terms of accuracy, the audit did not identify anomalies in the 2018 voters' roll that affected a large percentage of registrants or were they concentrated amongst registrants of a particular area, gender or age," he said.

"While no voters' roll is perfect, a less rushed process would have allowed more time for ZEC to identify and address anomalies.

"The 2018 voters' roll is more current than the 2013 preliminary voters' roll as there is a significant number of new registrants, as well as more urban and young registrants and fewer extremely old registrants.

"In terms of completeness, the 2018 voters' roll is more inclusive than the 2013 preliminary voters' roll which was with generally higher registration rates - though registration rates for urban and young voters remain lower than those for rural and older voters."

Mr Makoni said ZESN interviewed 1 518 individuals in its people-to-list test. ZESN also said the data of the 5 683 936 registered voters was accurate in terms of identifying information that was captured during registration.

"All 5 683 936 (100 percent) registrants have all six pieces of identifying information and all seven pieces of information specifying the registrant's polling station," said Mr Makoni.

"Two registrants have national ID numbers that are too long (15 rather than 14 characters). There is one registrant who will not be 18 by 30 July 2018, three born in the 1800's, 944 registrants who are 100 years old or older. Eighty-one registrants (less than 0,01 percent) have duplicate national ID numbers. Four thousand six hundred and ninety-three (0,8 percent) registrants have the same surname, forenames, gender, and date of birth as another registrant."

Mr Makoni said of the 3 213 780 matched registrants on both the 2013 preliminary voters' roll and 2018 voters' roll, 126 689 (4 percent) had different dates of birth, 107 603 (3 percent) had different surnames, 29 456 (0.9 percent) had different genders on the two voters rolls.

"The overall registration rate fell from 86 percent for the 2013 preliminary voters' roll to 79 percent for the 2018 voters' roll," he said.

"However, this was to be expected as the 2018 biometric voter registration created an entirely new voters' roll, thereby removing any suspect registrants from the previous voters' roll."