The Ripple Effect: the UBOS errors in the census and their wider implications

By Sebaggala Richard

The integrity of statistical data is critical to the design of effective public policy and the management of socio-economic development. As a student of economics and statistics, I am increasingly aware of the profound impact that mismanagement of data can have on governance and societal well-being. The recent controversy with the Uganda National Bureau of Statistics (UBOS) and its handling of the 2024 National Population and Housing Census data is a good example of this. UBOS was forced to recall the census report following widespread public outcry over inconsistencies in the population figures. The Executive Director of UBOS admitted that the demographic data had been misattributed, with population figures of different ethnic groups being incorrectly assigned to others. This situation raises significant concerns about the reliability of data collection, and processes, the accountability of statistical institutions, and the wider implications for policy making and social justice.

 

At the centre of the problem is the principle of accuracy. Statistical data is not just a collection of numbers; it represents realities that influence decisions that affect the lives of millions of people. When figures that are supposed to represent the Langi people are wrongly attributed to the Acholi, or when Acholi data is wrongly attributed to the Bakiga, the consequences go beyond academic concerns. Such errors can lead to misguided policies, misallocation of resources, and an inability to address the specific needs of different communities. These misrepresentations can exacerbate social tensions and inequalities and further undermine public trust in government institutions and the data they produce.

Population and housing censuses play a critical role in the distribution of government funds, voter turnout, and the identification of social needs. When demographic data is inaccurate, it can lead to inequalities in the distribution of resources, especially for marginalized communities. The impact of such inaccuracies is particularly great in developing countries like Uganda, where resources are limited and the need for targeted interventions is urgent.

In Uganda, the public demand for accountability following the misreporting of census figures reflects a growing awareness of the importance of accurate data. People are increasingly realising that decisions based on statistical reports have a direct impact on their lives. The calls for the resignation of those responsible at UBOS show that society expects more transparency and accountability in data management. This incident also raises important questions about the systems in place to ensure the accuracy of the data, as well as the training and resources available to those responsible for collecting and analyzing the data.

In light of this situation and public outrage, I sought to understand whether similar problems have occurred elsewhere — not to excuse UBOS, but to find out whether such incidents are more common than we think and how other societies have dealt with them.  A quick review of the literature revealed that while the UBOS error raises legitimate concerns, it is not unique to Uganda. Data management errors have occurred in other countries, even with advanced data systems, sophisticated infrastructure, and well-trained and experienced staff. This shows the need for stricter data management rather than questioning the mandate of statistical organizations.

For example, despite its sophisticated systems, the U.S. Census Bureau had problems with accuracy during the 2020 census. The bureau acknowledged errors in the count, particularly affecting racial and ethnic groups and impacting congressional representation and federal funding. These errors have led to reforms to avoid similar errors in the future. Although the context is different, the US case shows that even established statistical agencies with sophisticated systems can face data problems. The U.S. Census Bureau introduced a sophisticated privacy algorithm to protect individual privacy, but this led to discrepancies in the population counts of minority groups and rural areas. This demonstrates the importance of striking a balance between privacy and data accuracy (Mueller & Santos-Lozada, 2022). Even if the data error in UBOS is due to administrative oversight rather than complex algorithms, it shows the importance of sound data processing and quality control. UBOS, like other statistical organisations, must take this opportunity to review its processes and adopt global best practises to avoid future errors.

 

In the UK, an error in the National Health Service (NHS) patient record system in 2017 resulted in the misidentification of over 700,000 women for breast cancer screening. Thousands of women missed their screenings, with tragic consequences including undiagnosed cases of cancer. This error, caused by an administrative oversight, shook public confidence in the NHS, despite corrective action being taken. In addition, clinical coding errors in NHS hospitals between 2007 and 2010 led to incorrect payments for £1bn worth of treatment (O'Dowd, 2010).  Although these errors were primarily financial in nature, they demonstrate that administrative errors can have serious consequences—both for public confidence and the people involved. This emphasizes the importance of consistent monitoring and control in big data systems - a lesson that UBOS can learn from.

 

The Greek statistics scandal of 2010 is an excellent example of the far-reaching consequences associated with inaccurate data. The revelation that Greece had significantly underreported public debt and budget deficit figures to the European Union triggered a financial crisis and necessitated an international bailout. This incident is an important example for understanding the far-reaching consequences of statistical errors and shows how seemingly minor discrepancies can escalate into major economic disasters. Such events not only undermine public confidence but also have far-reaching consequences.

 

In response to the scandal, concerted efforts have been made across Europe to reassess statistical methodologies and reporting standards, while policy makers and economists have endeavored to mitigate the risk of similar incidents in the future. Although the error uncovered by the Uganda Bureau of Statistics (UBOS) may not have the same magnitude of impact, it demonstrates the extent to which data inaccuracies can jeopardize a country's credibility.

This incident emphasizes the need for a sound statistical framework and the ethical obligation of government institutions to ensure the accuracy of the data they disseminate. Promoting transparency, accountability, and ethical data practices is essential for statistical organizations. The establishment of independent oversight bodies tasked with reviewing and validating statistical practices is crucial. In addition, establishing mechanisms for public engagement and feedback prior to the publication of statistical results is crucial to prevent such unfavourable events from recurring.

In 2016, the Australian Bureau of Statistics (ABS) suffered a technical failure when its national census website crashed due to a cyber-attack. The incident, known as ‘CensusFail', caused delays and frustration. Although this was not a case of data transposition, it highlights how technical errors can affect data integrity and public perception. UBOS needs to communicate openly with the public to maintain trust and improve data accuracy, possibly through the use of new technologies.

These global examples are not meant to excuse UBOS but to show that even the most advanced statistical systems face challenges. Mistakes happen everywhere, regardless of the sophistication or experience of the institution. What matters most is the response— - admitting the error, correcting it, and improving processes to prevent it from happening again.

The Uganda Bureau of Statistics has over the years earned a good reputation for producing reliable statistical data. While this recent error should not be overlooked, it should serve as a catalyst for reviewing data verification processes and strengthening internal controls. Public trust is based on transparency. Therefore, addressing the problem openly and taking corrective action is essential to maintaining confidence in Uganda’s statistical systems.

In conclusion, I strongly believe that the technical staff and management of UBOS are taking proactive steps to resolve the problem. By acknowledging the error, working diligently to determine the cause, and committing to avoid similar errors in the future, UBOS can restore its integrity. By learning from this incident and implementing stronger internal controls, UBOS can strengthen its credibility and continue to play an important role in supporting informed decision-making, policy formulation and socio-economic development in Uganda.