INEC has now released results for the Bayelsa governorship election without holding elections in all polling units. INEC announced 352,552 votes or 71% for APC and 143,172 or 29% votes for PDP. These results are not consistent with the PVT estimates of between 62% and 46% for APC and 52% and 37% for PDP. This suggests that the results were manipulated during the collation process. If the tabulation process had been conducted properly then INEC’s official results would fall within the PVT estimates.
YIAGA AFRICA Parallel Vote Tabulation (PVT)
YIAGA AFRICA WTV adopts the Parallel Vote Tabulation (PVT) methodology for election day observation at the polling units. The PVT is a proven and advanced observation methodology that employs well-established statistical principles and utilizes sophisticated information technologies. YIAGA AFRICA WTV provides timely and accurate information on the conduct of accreditation, voting, and counting as well as independently verifies the official governorship results as announced by the INEC. PVTs are not exit polls. WTV citizen observers do not ask voters for whom they cast their ballot. Instead, the PVT relies on the official results from polling units for which there have been observers watching the entire process. This tested and proven election observation methodology has been deployed in over 50 countries around the world including Nigeria – most recently by YIAGA AFRICA for the 2019 presidential election.
YIAGA AFRICA PVT sample of polling units was drawn according to well established statistical principles and is truly representative of all of the polling units because the percentage of sampled polling units for each LGA is similar to the percentage of all polling units for each LGA. For example, Brass LGA has 8.80% of all the polling units in Bayelsa (159 of 1,804) and 8.80% (22 of 250) of the sampled polling units are in Brass LGA. While not identical, the percentages for every LGA are very close, clearly demonstrating that the PVT sample is representative of the entire state (Appendix I demonstrates the representativeness of sampled polling units).
High Percentage of Polling Units with no Election
The WTV data shows that election was not conducted in 24% (61 of 250) of YIAGA AFRICA’s WTV sampled polling units. This incident was more prevalent in 28 polling units in Southern Ijaw LGA, 11 polling units in Ogbia LGA, 9 polling units in Ekeremor LGA, 7 polling units in Nembe LGA, 5 in Sagbama LGA and 1 in Yenagoa LGA of the PVT sampled polling units. Statewide this data suggests that elections may not have been held in several hundred polling units thereby seriously compromising the rights of people of Bayelsa to freely determine for themselves their elected governor.
INEC Official Results
INEC has announced the official results for the Bayelsa election and made a return. As stated in YIAGA AFRICA preliminary statement, the PVT is not able to independently verify, regardless of the outcome, who won the Bayelsa gubernatorial election because elections did not occur in 61 of 250, or approximately 24%, of sampled polling units. This reduces the PVT sample and therefore YIAGA AFRICA cannot verify the outcome of the result.
However, the PVT is still able to determine if the collation process was conducted properly and expose manipulation during the collation process. If the official results as announced fall outside of the PVT estimated range, then the results have been manipulated during the collation process.
As Table 2 shows, the official results as announced by INEC for APC and PDP are not consistent with the PVT estimates, suggesting that the collation process may have been manipulated. INEC announced that APC received 352,552 or 71% of the vote while the PVT estimated range is between 62% and 46% meaning that the largest vote share APC could have received is 62% of the vote. Similarly, INEC announced that PDP received 143,172 or 29% of the vote while the PVT estimated range is between 52% and 37% meaning that the smallest vote share PDP could have received is 37%.
