The Sri Lanka DCS has been conducting the census of population and housing since 1871. Over the years the DCS has implemented new methods to improve its data collection. However, improving the quality and timely release of data has proven to be challenging. For instance, in the 2010 Census round, the DCS conducted the census in early 2012 and was only able to release the data in 2014. This time lag between the time of census and release of final report was due to delays inherent in mainly in using a paper-based questionnaire, then scanning the images to computerize the data. Errors in scanned images, and the time-consuming process for data analysis and validation posed further delays in the release of census data.
These challenges and consequent delays have affected the country’s development. Untimely release and poor quality of census data, delays policy decisions regarding fertility and mortality trends, migration, education, employment, disability and housing conditions. Population projections based on availability of such data are essential for planning public investments to ensure achievement of universal access to sexual and reproductive health and the SDGs overall.
Towards a Solution
To support the transition to Computer-Assisted Personal Interviewing (CAPI), at the request of DCS, UNFPA facilitated connection and collaboration with Vietnam’s GSO to initiate their experience sharing. The GSO in Vietnam has valuable experience in using CAPI for census as it had implemented CAPI using smartphones during its 2019 Population and housing Census.
With the support of the UNFPA country office in Sri Lanka, a delegation of 7 DCS staff (and 1 UNFPA staff) visited the Vietnam GSO in August 2019. The purpose of the 4-day (13 – 16 August 2019) study visit was to learn about Vietnam’s experience in the 2019 Census and the Survey on Women’s Health and Life Experience. The visit aims to help Sri Lanka apply this learning to improve the planning and implementation of its 2021 Census. The key elements shared by Vietnam during the visit included its experiences in (1) designing the 2019 Census with application of information technology (IT) in all stages of the Census; (2) developing and using CAPI software, web form, operation website and propaganda website for the Census; and (3) the National Survey on Women’s Health and Life Experience: Results of the 2010 Survey, methodology and the organization of the Second Survey.
This enabled the DCS team to understand and learn from the challenges faced by GSO in implementing CAPI. The lessons learned and recommendations shared by GSO focused on the importance of finalizing the questionnaire in advance in order to avoid making many revisions on CAPI, minimizing the number of software updates, providing guidance to enumerators on removing, installing and updating the CAPI application on their devices, conducting training classes with sufficient time for practical exercises, ensuring free storage capacity on devices to avoid losing data, data loss prevention, addressing data security issues, enhancing monitoring and supervising system. Based on these learnings, DCS was able assess its current implementation plan and introduce necessary adjustments. For instance, GSO had used the ‘bring your own device’ (BYOD) method which proved to be very challenging as the programming language of the CAPI application had to be modified to ensure compatibility with multiple operating systems. Learning from this, DCS was able to weigh the pros and cons and decide to proceed with using computer tablets as the most preferred modality. Further, DCS was able to strengthen the design of its system by making improvements to the data transferring process such as allocating time slots for enumerators to transfer data to respective supervisors, monitoring progress of field work using the Master Registry of census block database. With this new insight from Vietnam’s GSO, Sri Lanka’s DCS is better prepared for its 2021 census.
Both offices acknowledged the support from UNFPA which has been providing technical and financial support to NSOs. Its support to DCS to ensure that the census is of high quality, upholds international principles and standards, and produces data that can be widely disseminated and utilized for development was highly appreciated.