A digital census

March 12, 2023

Given the advantages of using handheld electronic devices, there are some challenges associated with these

A digital census


he seventh population and housing census of Pakistan has commenced. The self-enumeration portal is currently operative. Within a few days of its opening, the internet portal had recorded 4 million clicks. The survey’s fieldwork will be conducted by up to 121,000 field enumerators, who will use electronic devices to verify respondents’ information. This operation is scheduled to conclude by April 1. According to the Pakistan Bureau of Statistics, the data collected through this process will be utilised for public policy formulation and constituency delimitation in the future general elections. The information will also help the government in allocating resources more effectively in various areas. For example, areas with larger populations requiring additional medical facilities may be able to receive adequate funding and resources.

The NADRA has provided a comprehensive IT solution for the census. This is the first time such a solution has been implemented in the country. The census will take place across the entire country, covering 628 tehsils and approximately 185,000 census blocks. Android-based smart devices equipped with house listing and enumeration apps synced with GPS and GIS technology will be used to carry out the census. The Android-based digital census application supports both online and offline use and data synchronisation is easy. The system will support all three stages of the process - pre-census, census stage and post-census - for the Pakistan Bureau of Statistics. The NADRA is PBS’s technological partner for the initial digital census project.

In similar efforts, Iran spent two years designing and planning the 2016 electronic census with the help of 20 dedicated workgroups and committees. The 2017 e-census in Egypt took two years to organise, prepare and launch the pilot phase.

A strong planning phase is crucial for a successful census. In 2021, a technical committee was established for the census. It was asked to furnish its proposals for the upcoming census in three months. The PBS has already made the suggestions public. However, no action plan has been made public.

According to the United Nations Statistical Commission, a digital census is defined as one that uses satellite data, aerial photography, a geo-referenced address registry, and geographic information systems (GIS) for enumeration maps. These tools are essential for census mapping and therefore, the automation of the census (e-census/ digital census) would involve the use of digital maps at all scales during the pre-enumeration phase to demarcate enumeration regions. Widely distributing these maps to the general public and political leadership would greatly increase everyone’s confidence that their domains are adequately covered.

While the PBS plans to use a handheld data collection instrument for in-person interviews, other steps in the data collection process, such as data editing and cleaning, require a trained team and a reliable internet network to enable direct transfer of data from the handheld instruments to the central data centre.

It is suggested that the National Statistical and Spatial Data System can provide a map-digitising capability to enhance accuracy and enable visualisation of enumeration areas using imagery. Making these maps available to the general public can alleviate concerns that some areas of the country may be inadvertently overlooked during the census.

The use of computers rather than a paper questionnaire incurs expenses and hazards. Using handheld devices for data collection poses obstacles such as high equipment costs (with limited long-term use) and longer preparation times.

Census data collection is undergoing a revolutionary transformation due to the rapid advancements in information and communication technologies. The integration of these new methods has immense potential for enhancing national statistical data collection programmes. Recent advancements in handheld computing devices and mobile connectivity have given rise to innovative approaches for collecting data that are not only faster but also of higher quality than traditional paper-based methods. Handheld electronic devices such as tablets, PCs, cell phones and personal digital assistants have demonstrated the ability to expedite data collection and improve data quality. Even in areas with limited energy and mobile network infrastructure and challenged institutional capability, these technologies have the potential to enhance data quality and timeliness, making them the standard in field-based data collection.

In particular, computer-assisted personal interviewing has replaced the conventional approach of using paper and pencil for data collection.

The use of portable electronic devices offers many advantages in terms of efficacy, efficiency and data quality compared to traditional paper-based methods. Rapid digitisation of data at the site of collection eliminates the need for transcription to a centralised computer system and enables automated and quick data aggregation. Fieldwork errors can be remedied and modifications to methods can be made rapidly when data are available for inspection by field supervisors and data processing teams. Software guidance improves the interview experience for both the interviewer and respondent, ensuring greater data quality.

Despite the advantages of using handheld electronic devices, there are also some challenges as well. Using handheld devices for extensive data collection activities poses significant obstacles, such as high equipment costs with limited long-term use, longer preparation time and the need for sufficient technical expertise and support. There are also concerns about physical safety and security, privacy and confidentiality and infrastructure constraints. Online data gathering poses difficulties and risks, particularly for large-scale statistical operations like population and housing censuses, which are among the most complex and comprehensive data collection operations.

The integration of internet-based options with traditional methods of collecting and distributing census forms is feasible. However, it is crucial to ensure that each household and individual is accurately counted only once, especially when employing a multi-modal approach. This strategy presents several challenges, including the need for a reliable connection between each household and its geographic location, coordination of multi-mode collection operations and the provision of adequate systems infrastructure. Additional difficulties include ensuring confidentiality and security of data, accommodating an extended collection period that could affect the accuracy of data captured, the possibility of introducing bias due to nonresponse or mode effects and the high initial cost involved.

Japan conducted a census in 2015 employing a multi-modal approach. Households were provided with three different ways to respond to the census, including self-enumeration using a paper questionnaire, online responses or collection of data by enumerators. Of all the households surveyed, 36.9 percent opted to complete the online questionnaire. This approach demonstrates the feasibility of integrating multiple data collection options and the potential benefits of employing a multi-modal approach to achieve accurate data.

There should be a comprehensive evaluation of IT investment management once the census project is implemented. The evaluation is essential to validate the actual investment results and compare them against the set estimates and expectations to assess performance. Moreover, the lessons learned from the evaluation process can be utilised to enhance future IT investment decision-making and implementation processes. The evaluation phase is critical for improving the way an organisation selects, manages and utilises its IT resources.

The author is experienced researcher, writer and analyst in field of cyber security, he has LLB, M.Phil in Cyber Crimes and currently pursuing his Ph.D in CS

A digital census