PRELIMINARY STATEMENT ON GHANA’S PRESIDENTIAL ELECTION RUN-OFF HELD ON SUNDAY, DECEMBER 28 2008

INTRODUCTION

On Sunday, December 28, 2008, the Coalition of Domestic Election Observers (CODEO), which comprises thirty-four (34) professional, religious and civic advocacy bodies (including the Christian Council of Ghana, the Federation of Muslim Councils, the Ghana National Association of Teachers, the Trade Union Congress, and the Ghana Federation of the Disabled), deployed about   4000 trained and accredited non-partisan domestic election observers throughout the country. Among the 4000 observers were over 1000 Rapid Response Observers (RROs) deployed to a nationally representative sample of polling stations in all the 230 constituencies.

This preliminary statement reflects CODEO’s observations on the conduct of the presidential run-off election. It incorporates some of the observations noted in the two statements issued by CODEO on the election-day regarding the setting-up, opening and closing of polls and the overall voting process on Sunday, December 28 and on Monday, December 29, 2008 respectively.

CODEO has received reports from nearly all (99%) of its RROs (with the exception of observers deployed to the Tain constituency (in the Brong-Ahafo region), which did not participate in the polls on December 28. This report also contains data received from some of our regular observers as well as reports filed by CODEO Regional Coordinators (RCs) and Constituency Observers (CSs).

SUMMARY OF OBSERVATION

Preliminary analysis of reports filed by CODEO observers suggest that the December 28 2008 presidential run-off was conducted largely in accordance with the electoral laws of Ghana and relatively few problems and lapses were encountered in the set up and opening of polls.  However, there was relatively more intimidation and violence in the voting process this time around.

POLLING PROCESS OBSERVATION

CODEO observers were permitted to observe the setting-up and the polling process in nearly all polling stations (99%). Nearly all of CODEO observers (98%) had reported to their assigned polling stations by 7am. And most of the polling officials had reported at the polling stations at the time of the arrival of CODEO observers. 

Setting Up and Opening of Polling Stations

• On the whole, the set-up of polling stations and opening of polls was significantly improved, compared to December 7 polls. Relatively few problems and lapses in procedure were recorded by our observers. Unlike on December 7 when about a third of the polling stations did not open on time (i.e. 7am), majority of the polling stations (79%) opened at about 7am. In the few places where opening of stations was late, voting often started before 9am. 

• EC officials were able to transport election materials to polling stations in a timely manner. All election materials were present at the time of setting-up in the vast majority of polling stations (97%). There was a sufficient number of presidential ballot papers in most polling stations. 

• The agents of the two contesting parties in the run-off (the New Patriotic Party (NPP) and the National Democratic Congress (NDC)) were well represented at the polling stations (97% for the NPP and 99% for the NDC). 

• Uniformed security was present at most polling stations (86%). This compares with 88% in the December 7 polls. CODEO observers also noted the presence of mobile security forces who tended to respond to reported incidents promptly. In the Northern region, however, uniformed security personnel were less visible at polling stations (3 out of 10 polling stations). 

• Nearly all polling stations (98%) observed were set up so voters could mark their ballots in secrecy. About 98% of polling stations were accessible to persons with disability and the elderly which was an improvement on the December 7 figure of 95%. Also, ballot boxes were shown to be empty, sealed and placed in the public view at polling stations. 

Voting Process

EC officials largely adhered to the rules governing voting, although our observers reported some disorder and disruptions in the process. Reports filed by our observers generally indicated the following: 

• Voters with valid voter IDs cards were permitted to vote in most polling stations (83%). In 88% of the polling stations voters had their names checked in the voters’ register before being allowed to cast their ballot. 

• In a few polling stations (5%) voters were allowed to vote without voter ID cards. This particular incident was recorded in all the ten regions, with the most coming from Western, Volta, Central, Ashanti and Eastern regions. This represented a marginal increase from December 7 (4%). While this is permissible under EC regulations, CODEO was unable to confirm whether the right procedure was followed by the polling officials by allowing these voters to cast their ballots. 

• CODEO observers reported few cases (69%) of disabled and elderly voters being assisted to vote. 

• In most polling stations (98%), voters had their finger marked with the indelible ink before they cast their ballot.  However, our mobile observers recorded instances in the Volta Region where the indelible ink was applied to the left finger after voting. 

• There were few instances (3%) where ballot papers were not validated before being handed to voters.  

• There were fewer voters (10%) waiting in queues to cast their ballot at the official close of polls (compared to December 7 when it was close to 30%). In 89% of polling stations there were no queues by the time voting officially ended.  

Counting and Collation of Votes

• Counting and collation of ballots were generally orderly. In most cases the number of ballot papers tallied with the number of votes cast. 

• Party agents requested ballot recount in nearly 1 in 10 of the polling stations observed (9%).  

• Party agents refused to sign declaration forms in 5% of the polling stations observed by CODEO 

• In nearly all instances (99%) CODEO observers (99%) agreed with the vote count in the polling stations where they were deployed. 

Voting Irregularities and Incidents

In general, the presidential run-off recorded relatively more incidents of voting irregularities in the run-off  than the December 7 elections. Whereas 134 cases of incidents were reported in the December 7 election, the run-off had 201 cases. This was most pronounced in Brong Ahafo, Eastern and Ashanti regions, which witnessed increases in reported number of incidents (i.e. +42, +38 and +35 respectively). The number of incidents The Western and Central regions recorded dramatic reduction in incidents (see Table 1). 

Table 1: Cumulative Number of Reported Incidents (Run-off and Dec. 7 Compared)
	DEC. 7 Election	DEC. 28 Run-off	Difference
Brong Ahafo	11	53	+42
Eastern	10	48	+38
Ashanti	8	43	+35
Central	22	12	-10
Greater Accra	17	15	-2
Northern	15	6	-9
Upper East	15	9	-6
Upper West	3	2	-1
Volta	12	8	-4
Western	21	5	-16
Total	134	201	67
Note: The difference is the reported number of incidents on December 7 less the Run-off figures.

• CODEO Constituency Supervisors, Regional Coordinators and roving  observers reported incidents of f unauthorized road blocks mounted and  on-the-spot checks conducted on vehicles and persons by militant  vigilante groups in parts of Anloga and Ave Avenor constituencies in the Volta region.. In some instances, the mob prevented ‘so called’ strangers from entering the constituency and/or polling stations ostensibly to counteract the rumoured presence of ‘macho men’ allegedly roaming  around the area to snatch ballot boxes.  At Anloga observers encountered a mob of local vigilantes confronting the police and preventing access to the police station. The Police told the observers that NPP agents have sought refugee at the Police station after a couple of their colleagues had been beaten up. 

• Majority of the incidents reported by our observers (39 cases) occurred in the course of the setting up of polling stations (a marginal reduction from the 42 cases reported during December 7 elections). The incidents at the set up were mainly related to missing election materials such as ballot boxes, ballot papers, voters’ register, indelible ink, voting screen, validating stamp, endorsing ink and ink pad. Most of these problems occurred in parts of Brong Ahafo, Eastern, Ashanti, Upper East and Volta regions. Almost all of these problems, according to our observers, were resolved before voting commenced. 

• Our observers also reported 30 cases of disorder at polling stations compared to 7 cases on December 7. These incidents occurred mainly in parts of Eastern, Ashanti, Central, Volta and Greater Accra regions. Such incidents might explain the increase in cases of suspended voting from 7 in the December 7 polls to 11 in the run-off.  Indeed, the reported cases of suspended voting occurred mainly in parts of these same regions. For instance, suspension of polls was reported in parts of Ashanti (Temp Booth Freeman Hotel polling station in Asokwa constituency), Central (Methodist Church/Town Saltpond polling station in the Mfantseman West constituency, CPB Amantsiaa Abaasa polling station in the Ajumako/Enyan/Esiam constituency), Eastern (L A JSS Anyankode Awukugua polling station in the Okere constituency) and Greater Accra (Oblogo in Weija constituency). Such incidents were also reported in Wli-Agorviefe and Wli-Afegave polling stations in the Hohoe constituency in the Volta region. 

• Our observers reported that voting never took place in the Tain constituency in Brong Ahafo because there were shortages in materials distributed to polling stations in the constituency and agents of both parties opposed EC’s proposal to do a recall and redistribution. 

• Once again, voters in Temp Booth Chalator polling station in Kpandai constituency in the Northern region refused to vote just as they did on December 7. 

• Incidents of intimidation and harassment also increased from 5 cases during December 7 elections to 18 in the run-off. Similarly, the number of cases of violence went up from 4 during December 7 elections to 18 in the run-off. Perhaps, the two types of incidents are linked since Ashanti, Brong Ahafo and Eastern regions reported high number of cases of both. CODEO RROs were not present at the places in Volta Region where similar anecdotal reports of such incidents have come out 

• There was no significant change in the reported number of cases of violation of voting procedures (6 cases in the run-off and 5 cases during December 7). Most of the reports relating to this incident are from parts of Ashanti and Northern regions.  However, our roving observers reported a few cases in the Volta Region where the indelible ink was applied post voting) 

• CODEO observers reported a few cases of ineligible voters being allowed to vote; disenfranchising of qualified  voters; ballot boxes not shown to be empty and sealed; stealing of ballot boxes; polling officials being suspected to be partisan; stuffing of ballot boxes; and other incidents such attempted multiple voting, some party agents leaving polling stations with explanation, some individuals claiming to be party agents yet not having accreditation and disagreements between party agents and electoral officials (see Figure 1). 

Figure 1: Minor Incidents Reported by Observers (Run-off and Dec. 7 Compared)

 

PARALLEL VOTE TABULATION (PVT)

As part of its overall effort to observe the 2008 elections, the Coalition of Domestic Election Observers (CODEO) conducted Parallel Vote Tabulation (PVT) in the December 7 and 28, 2008 Elections. CODEO deployed trained accredited Rapid Response Observers (RROs) to randomly selected nationally representative sample of over 1000 polling stations from the official list of polling stations provided by the Electoral Commission (EC). For good representation, the over 1,000 polling stations were first allocated to the ten regions according to their percentage share in the total polling stations in the country (see Table 2 for the regional distribution). 

Table 2: Distribution of Sample Polling Stations by Region
Region	Polling Station by Region	% Share in Total Polling Stations	PVT  Polling Stations per Region	% Share in Total PVT Polling Stations
Ashanti	3,667	17.5	185	17.3
Brong Ahafo	2,382	11.3	120	11.2
Central	1,763	8.4	93	8.7
Eastern	2,609	12.4	135	12.6
Greater Accra	2,457	11.7	125	11.7
Northern	2,043	9.7	103	9.6
Upper East	1,028	4.9	50	4.7
Upper West	854	4.1	44	4.1
Volta	1,965	9.4	100	9.3
Western	2,240	10.7	115	10.7
Total	21,008	100.0	1,070	100.0
Note: Number of polling stations provided to CODEO by the Electoral Commission.

 The number of polling stations allocated to any region was then distributed across the constituencies within that region based on their percentage share in the total polling stations in the region (see Figure 2 for the distribution of polling stations in Volta). 

Figure 2: Distribution of Sample Polling Stations across Constituencies in Volta Region
Note: Number of polling stations provided to CODEO by the Electoral Commission.

 Field data gathered by RROs were sent directly into CODEO computerized database at its Observation Center at the Kofi Annan International Peacekeeping Training Centre (KAIPTC) by means of special text messages.

 

 PVT Estimates of December 7, 2008 Presidential Election

CODEO indicated in various pre-election press statements that the inclusion of the PVT technique in its election observation toolkit for this year’s election was to help verify scientifically the accuracy of the result of the presidential election results as announced by the Electoral Commission. Indeed, by 2:00pm on Monday December 8, CODEO had been able to estimate the outcome of the presidential election based on information received from 99.8 percent of RROs. On Wednesday, December 10, CODEO in a press conference at the International Press Center released the PVT estimates of the December 7 Presidential election and noted the following: 

• the percentage of the vote received by the candidates was consistent with the official figures released by the EC;  

• the order of candidates based on their performance in the elections in the PVT matches that of the official results; and 

• the official result for each candidate fell within the PVT estimated 95 percent confidence interval given the estimated margin of error for each candidate (see Table 3 for PVT estimates). 

Table 3: PVT Estimates of December 7, 2008 Presidential Election (% vote received)
Candidate	Political Party	PVT Margin of Error	PVT Estimated Results (%)	EC Official Results (%)
Nana Addo Dankwa Akufo-Addo	NPP	±1.6	49.8	49.1
Prof. John Evans Atta Mills	NDC	±1.6	47.4	47.9
Dr. Paa Kwesi Nduom	CPP	±0.1	1.3	1.3
Dr. Edward Nasigrie Mahama	PNC	±0.2	0.8	0.9
Emmanuel Ansah-Antwi	DFP	±0.0	0.3	0.3
Kwesi Amoafo-Yeboah	Independent	±0.1	0.2	0.2
Kwamena Adjei	RDP	±0.0	0.1	0.1
Thomas Ward Brew	DPP	±0.0	0.1	0.1
Note: PVT estimate range is based on a 95% confidence level

  

• The coalition also observed that the PVT estimated voter turnout of 70 percent and rejected ballots of 2.2 percent closely matched the official figures of 69.5 percent and 2.4 percent respectively. Given the close convergence of the PVT estimates and the official results, CODEO confirmed EC’s position that there was no outright winner. 

PVT FINDINGS ON THE RUN-OFF PRESIDENTIAL ELECTION

Ladies and gentlemen, in fulfillment of the promise made during the 10^th December press conference, CODEO has invited you to present its PVT findings on the Presidential run-off between Nana Addo Dankwa Akufo-Addo (NPP) and John Evans Atta Mills (NDC).

 As of 8:00am on Monday, 29th December, CODEO had received reports from 1,062 RRO polling stations constituting 99 percent of the PVT sample. This excluded those from Tain Constituency in the Brong Ahafo region where voting never took place and Temp Booth Chalator polling station in Kpandai Constituency in the Northern region where electorates refused to vote. The total number of valid votes recorded in these polling stations in the December 28 polls is 456,728 (compared to 448,624 on December 7).  

 

 Our PVT estimates show that Nana Addo Dankwa Akufo-Addo, the NPP candidate received 49.8 percent of the valid votes cast while NDC’s Prof. John Evans Atta Mills received 50.2 percent (see Table 4). 

 

Table 4: PVT Estimates Presidential Election (% vote received)
Candidate	Political Party	PVT Margin of Error	PVT Estimated Results (%)	EC Official Results (%)
Nana Addo Dankwa Akufo-Addo	NPP	±1.7	49.81	49.87
Prof. John Evans Atta Mills	NDC	±1.7	50.19	50.13

Our estimate of the voter turnout which is 72.3 percent is a 2.3 percentage point increase over the December 7 figure of 70 percent. This figure matches closely with the official figure of 72.74 percent declared by the EC. Also, PVT estimate of rejected ballots (1 percent), which is an improvement on the 2.2 percent estimated for the December 7 election is not significantly different from EC’s figure of 1.02 percent. 

CODEO can confidently confirm that the results of the PVT are consistent with the official results of the presidential run-off announced by the Electoral Commission. 

Political contestants, their supporters and the general public should therefore have confidence in the accuracy of the official results declared by the Electoral Commission on December 30, 2008. 

GENERAL CONCLUSIONS

CODEO reiterates its earlier position in the statement issued on Monday, December 29 2008 the problems and lapses in the voting process in the December 28 presidential run-off do not fundamentally undermine the integrity of the overall process. Thus, CODEO concludes,  in this preliminary statement, that the December 28 presidential run-off had been generally well conducted. 

Accordingly, CODEO commends Ghanaians for turning out in their numbers to express their political will and affirm their fundamental right to choose who should become their president for the next four years. 

CODEO has also taken note of the significant reduction in spoilt ballot papers during this run-off and commends those responsible for educating the public on how to vote properly. 

The Coalition commends the leadership and staff of the Electoral Commission (EC) for the able manner in which they conducted the run-off. It also commends the security agencies for their professional conduct and swiftness in responding to various incidents of disruption and violence. We urge the Police Service and the other security agencies not to relent in their efforts to maintain peace and security in this transition phase and beyond. CODEO recommends that the Police Service takes appropriate steps to initiate the prosecution of all those arrested for committing electoral offences. 

Ladies and Gentlemen, CODEO will observe the Tain Constituency polls and continue to monitor the post-election environment. A detailed report with recommendations for further improvements in our electoral process will published in early 2009. 

In the meantime, we call on all Ghanaians, including leaders of political parties and their supporters, to continue to place full confidence in the Electoral Commission.  

We also encourage the two presidential candidates and their respective political parties to accept the final official results. We call on the candidate who loses the Tain constituency election scheduled for Friday to gracefully concede defeat, and the victorious candidate to be magnanimous. 

CODEO notes with concern incidents of intimidation that were amplified via incendiary statements and commentaries broadcast on some radio stations. We urge media outlets and all Ghanaians to refrain from divisive   pronouncements and actions to ensure that the post election atmosphere remains peaceful. 

The president-elect, when sworn, in will be President of all Ghanaians in the coming four years and must therefore move quickly beyond the partisan politics of the electoral cycle to earn the support of all Ghanaians. 

Finally, CODEO appeals to all to remain calm as the nation prepares for yet another important transition. 

Thank you and God Bless Ghana 

Signed:

Professor Miranda Greenstreet and Justice VCRAC Crabbe,

Joint Chairpersons

(For and on Behalf of the Advisory Board)

CODEO Secretariat, Wednesday, December 31, 2008

 

 

 

Uncategorized