Cameroon Biometric Voters Registration (a Gold Mine or electoral innovation?)

Cameroon Biometric Voters Registration:

A Gold mine or an electoral innovation?

 In a number of my articles, I cautioned that while the nation may have made the right decision to go biometric, self center individuals, institutions and organizations may frustrate the process and bring the nation into another round of bitterness and finger pointing instead of uniting towards the meaningful construction of our democracy. Mindful of the amounts involved in this whole process, the civil society and the CONAC must be on the high alert and probe into some of the processes.  The probe is to further strengthen the electoral process and give voters renewed confidence in the electoral process, uncover irregularities and solve them early enough before they generate into an incurable virus in the whole system. Such a process will begin with the system definition, the terms of reference and the processes involved in issuing the calls for tender and analyses of the bids.  All potential bidders should have the same access to information in form and time. Go pass that point, all five companies shortlisted and who demonstrated their know-how should be double checked against the facts they presented and the track records they are hanging on. I understand some of these companies said they have track records in  Ghana,  Mali, Bangladesh, Tanzania, Zanzibar, Guinea, Rwanda, Swaziland, DRC, Malawi, Nigeria, Kenya, Gambia, Iraq, Haiti, Uganda, Lesotho etc

But reading through the net, articles such as these  (http://www.unpo.org/article/2983) are not the things we will like to see about anyone trying to supply such a very sensitive system to our country.

A biometric registration process is a multilevel process starting from the most crucial which is the software, the hardware, the training, the data collection, card production and then the top priority; the compilation (which include matching, adjudication, ) and the final production of voters list. Having a picture and a finger print that is not digitally analyzed on a voters card is not biometric and must not pretend to be so. 

This article that appeared in an online news outlet cameroon-info.net (http://www.cameroon-info.net/stories/0,33004,@,marche-de-la-biometrie-sani-tanimou-veut-imposer-les-sud-africains.html) gives reason why we must as soon as possible keep the records straight and move the electoral process forward.

Drawing from an excellent publication by the Electoral Institute for Sustainable Democracy in Africa (EISA http://www.eisa.org.za/PDF/vrafrica.pdf), There are many countries that use, or have used, biometric voter registration technology, but if the conditions are not right, it can be far from successful. Several experts stress the need to carry out a thorough needs assessment and feasibility study before deciding to introduce biometric technology in either voter registration or verification. The various challenges and the time required to implement such technology should not be underestimated.

Some of the particular challenges associated with the introduction of voter registration biometric technology are highlighted from Uganda’s experience: time required registering voters; the calibration of the facial recognition software; and difficulty of obtaining finger prints from some people.

Testing biometric systems

It may be difficult for our election Management Body (EMBs) to analyse and fully understand study results presented by vendors who are eager to sell their products. Simple studies are good for obtaining a rough overview of a system’s performance; however, more in-depth studies are crucial before a system is deployed to ensure that real performance will meet commission’s requirements and that the system is fully operational in specific country conditions.

Comparisons of systems are difficult since the methodologies used to conduct the studies or the way errors are calculated can vary enormously. Vendors’ calculations of a system’s false reject and false accept rates often rely on research studies or small test applications of the system. Case studies  conducted in an air-conditioned offices using college students are not a realistic representation of how the voter registration system would work in extreme Cameroon conditions.

For example, the scanners purchased by Ghana to process OMR forms quoted a processing speed of 5,000 forms an hour, while in practice the ‘real’ speed was some 1,000 forms an hour.

Understanding the basic principles and asking direct questions is the best way to ensure that the system chosen meets a country’s needs. For the most accurate results, the commission should insist that research studies be conducted in the field under ‘real’ conditions, or at least in a way that strongly simulates real-world conditions.

In addition, the sample group should reflect a cross-section of voters’ ages, professions and skin conditions. For example, the sample group should include senior citizens, smokers, gardeners, bricklayers, etc. and people whose fingers are, for example, very dry, broken, moist, etc. The environmental conditions must include testing under extreme high and low lighting conditions, and under different humidity and dust conditions.

Iris and face recognition technology

Iris and face recognition technology work on a similar principle to fingerprint recognition. Iris recognition involves recording the image of an iris (coloured part of the human eye) using a high-resolution digital camera. The software mathematically analyses the pattern of the iris and converts it into a 512 byte digital template that is stored in the database for future reference. This technology is considered to offer the highest accuracy in capturing biometric data because no two human irises are alike. In addition, the iris is very stable: it does not change with age and is less influenced by environmental conditions. Iris recognition does not involve retina scanning (which is a completely

Different technology) and is therefore not invasive and unsafe. A picture of the eye can be taken without any personal contact, and contact lenses or glasses do not interfere with the accuracy of the image.

Biometrics – the black box approach

Does biometrics really improve the electorate’s trust in the voter registration system or does the ‘black box approach’ give electoral stakeholders decreased control and little room for observation? In Mozambique, for example, several stakeholders accused the National Electoral Commission of lack of transparency because the commission refused to disclose details of the computer processes used in the voter registration exercise and compilation of the voters’ roll, and would not release data necessary to identify the number of voters registered at each voting station. In Senegal, even staff at the ministry in charge of the database considered the entire IT component of the data processing system (which is managed by external commercial companies) as a black box, with no real possibility to supervise or effectively audit the process.

Enfranchisement of voters lost in technology

The use of high-level technology in voter registration has raised the issue of moving fragile electronic equipment across treacherous terrain, sometimes resulting in the loss or corruption of the very information that is being sought after. These days it has become difficult to register voters close to their homes (polling station level) resulting in confusion among voters when it comes to locating their proper voting stations on polling day. The once good precept ‘where you register is where you vote’ has been challenged by technology.

It is important to understand and evaluate the messages sent through the use of different technological approaches. Voter inclusion and the enfranchisement of disadvantaged groups (such as women or people living in remote areas) have to be balanced against the security features of a system that prevent double registration by technological means.

Practice shows that other transparency features of a system are important to the process. These include the display of the voters’ roll at local level, the provision of sufficient time for the engagement of political parties, CSOs, and the involvement of citizens in the process. Intense civic and voter education raises awareness that a functioning democracy relies on the participation of each citizen and that a system of checks and controls is best implemented at grassroots level.

Key findings of the case studies

Key findings of the case studies are listed below: while some state the obvious, they collectively provide an overview of the issues and lessons learned.

-        Political reality determines the necessary security features of a system.

-        Inclusiveness in planning and the evaluation of voter registration processes is needed to maximise system performance.

-        Confidence in the voters’ register is dependent on confidence in the impartiality and independent decision-making of the commission.

-        The best system designed will fail if it is not managed properly.

-        Transparency of the system, regular communication and stakeholder participation is crucial for acceptance of the process.

-        Efficient timing is critical: a delay in voter registration can jeopardise the whole electoral process; lead times for equipment purchase need to factor in potential delays in supply.

-        A balance is needed between providing enough time for data processing and cleaning, and holding voter registration close to the election date.

-        Systems need to be tested rigorously before being implemented nationally.

-        Using a smaller number of well trained and well equipped field data collection teams in a staggered process is easy to manage and can ensure good coverage of remote areas.

-        Training for a computer-based system has to be completed before the system is implemented and must fully account for participants’ existing skills levels.

-        Developing sufficient in-house technical skills to maintain a voter registration system is essential to sustain its reliability and integrity. However, it is difficult for the commission to hold on to skilled IT personnel as they are likely to accept jobs in the private sector or with international organisations. Even when they stay, clear job definitions are needed. The current ELECAM website (www.elecam.cm) continue to talk about presidential election many months after the end of the polls and completely nothing about the current processes.

-        Field equipment should be simpler and more robust the harsher the conditions in the field.

-        There seems to be a negative correlation between the level of technology and the number of registration points offered to voters.

-        High-tech systems cannot solve political issues or problems of trust in the overall process and can jeopardise sustainability issues.

-        High-tech equipment tends to get stolen or damaged.

-        Biometric data collection is difficult and expensive relative to the benefits that can be gained from de-duplication.

-        The reliability of fingerprint matching as a security feature has not yet been proven.

-        Providing a photo voter’s card can be a strong incentive for registration.

-        Computer equipment requires a realistic storage and maintenance plan.

-        Once a high-tech system is in place, it is difficult to revert to a simpler system.

-        Display of the voters’ list and ensuring there is enough time for active citizen participation increases acceptance of the voters’ roll.

-        It is very difficult to improve an existing voters’ roll; rather scrap it and start a new one from scratch.

-        Printing photographs on the voters’ roll is a high-level security feature to prevent impersonation.

-        Capacity for equipment storage and maintenance is crucial for the sustainability of a system.

-        Continuous registration is not cost effective if it is not supported by continuous voter education.

-        The commission cannot base the voters’ roll on information gathered by other state sources unless state civil registration is very effective and requests citizens to notify the authorities of any changes of address.

-        Voter registration methods, processes and management structures may be an appropriate basis for the development of a civil registry.

-        Voter integrity controls have to be appropriate for the commission’s capacities and for external information availability, and where appropriate should be easily publicly accessible.

-       Transparency in voter registration operations is essential for building trust in the integrity of the electoral process.

Tanyi Christian E
LUKMEF-Cameroon
Website: www.lukmefcameroon.org
Mission: To Promote Peace, Nonviolence, Social Justice, Human Rights  and Sustainable Development in Cameroon