Auditing of voter registration is of prime interest to stakeholders of the electoral process. It provides the opportunity to understand the processes for establishing/updating voter registers as well as assessing the resulting voters register. Given the role played by voters registers in elections, several attempts have been made to standardize their auditing procedure. However, u to now, there is no systematic approach to their verification. This is in part due to the complex process leading to establishing or updating voters registers. The approach suggested to auditing voter registration is based on first assessing the process itself and later using statistical testing as a tool to collect auditing metrics (i.e. evidence) to enable interpretation of facts discovered when evaluating the registration process.

The electoral cycle approach provides three entry points for the audit of voter registration processes:

• Pre-electoral period: The quality of the existing voter register needs to be checked in the pre-electoral period as part of the planning and implementation of the electoral operations. This makes it easy for the electoral administration to update the voters’ records, if necessary. To ensure credibility in the electoral process, it is required that stakeholders are involved in the audit
• Electoral period: A second assessment should be done to address the objections raised by voters. This assessment should take place in each electoral district and registration centre and all objections need to be supported by evidence to eliminate subjective issues
• Post-electoral period: When the vote is over, the electoral cycle provides for audit and evaluation activities. In this case, the audit identifies possible improvements to the existing voter registration

The Control Objectives for Information and related Technology (COBIT) represents a set of rules that facilitate the audit of an organisation’s – such as an EMB in charge of the voter registration process – internal processes. The stages of an independent COBIT-based audit process are as follows:

1. The audit charter (approved by the EMB)
2. The independence clauses
3. The professional ethics and standards to follow
4. The competence of the auditor (the scope of his or her work)
5. The auditing plan
6. The audit per se
7. The report
8. The follow-up activities of the audited organisation

Audit of voter registration processes entails the audit of two elements, namely the registry itself but also the procedures for its creation and/or update. Three techniques are used to this end:

• Audit through a COBIT-adapted framework to evaluate and assess the registration process and thus the flow of information from data collection forms to the creation of access points for voters and all the way to the registration of voters (and, if applicable, the issuance of voters cards)
• Statistical testing to assess the quality of the voter register and the voters list through the “list to voters” and/or the “people to list” methods
• IT audit-standard practice to examine the technological components and infrastructure of the voter registration process and registry

Case Study: Voter Registration Audit in Yemen

The National Democratic Institute for International Affairs (NDI), in collaboration with the Yemen Electoral Monitoring Network (YEMN), carried out an audit of the voter registration process in Yemen in 2008. The audit was undertaken partly as a result of distrust among the local political parties regarding the impartiality of the electoral commission and partly because of the widespread protests that occurred during the registration process.

NDI and YEMN selected 13 governorates to monitor. The governorates chosen were particular in two respects: they either had disproportionately low levels of female representation in the registry and/or the number of registered voters exceeded the actual population. A 5% random sampling method was used to further select the registration centres to be monitored. In total, YEMN monitored 330 of 5620 centres (5.8%) in the 13 selected governorates.

Examples of the findings of the monitors include:

• That there was a lack of proper procedure in the registration • centres, including for example the failure to ask for identification documents, failure to check faces against identification documents, incorrect recording of voter information, registration of voters at unscheduled hours and the collection of voter information by community members and security personnel
• That registration centres did not receive adequate supplies (film, voter identification cards, cameras etc.) or the equipment was unusable. Moreover, in some instances the commissioners did not know how to use the Polaroid cameras that had been distributed by the electoral commission
• That community leaders and security personnel interfered with the work of the registration staff through for example pushing commissioners to register ineligible voters
• That the voter registration environment was marked by protests and in some instances violence or threats of violence in unstable areas