Operationalising the UNODC Statistical Framework to Measure Corruption

Efforts and approaches to measure corruption have made big strides over the last two decades73207abeca0a with the dual purpose of understanding an evolving phenomenon as well as measuring the effectiveness of anti-corruption efforts.

There has been a move beyond corruption perception surveys and rankings, which have had and still have a tremendous impact on awareness raising and advocacy but have been difficult to operationalise. To inform concrete reform needs and reliably monitor progress towards reform goals, actionable data is needed.

The measurement of Sustainable Development Goal (SDG) target 16.5 shows how challenging it is to even collect reliable and comparable data on a relatively simple indicator such as bribe payments within the last 12 months by households and businesses. Only 50% of 45 cross-national and national corruption surveys provide sufficient data to monitor the SDG target.becf826e1c4c

The Statistical Framework to Measure Corruption (2023) by the United Nations Office on Drugs and Crime (UNODC) was developed through consultations involving experts and institutions across more than 80 Member States and is way more ambitious in many regards, with the objective ‘to provide guidance for national governments to develop national information systems able to detect the presence, measure the magnitude, and monitor trends of the different forms of corruption, guided by the United Nations Convention against Corruption’.

In its latest version, the framework offers a menu of 145 indicators covering criminal offenses, preventive measures, and the broader enabling environment to report and address corruption offenses and risks under the UN Convention against Corruption (UNCAC). The framework distinguishes between risks and anti-corruption responses, in regulation (de jure) and implementation (de facto), for example, the existence of conflict-of-interest regulation (de jure) and the percentage of public officials sanctioned for not disclosing conflicts of interest in the previous year (de facto).

The two measures for SDG 16.5 are included, but international comparability is not the main aim of the framework. The framework deliberately encourages national adaptation and the selection of a set of the most relevant indicators in a particular context. This improves policy relevance but can reduce strict cross-country comparability.

The indicators draw on multiple kinds of evidence, including:

• Household and business surveys
• Administrative records
• Criminal justice and enforcement statistics
• Institutional and regulatory data,
• Data on preventive systems, reporting and complaints mechanisms
• Perception and contextual measures.

Data on some of these indicators may be readily available at relevant government agencies; some data may be collected by various bodies, albeit using different definitions; some data may not yet be collected at all.

A consequential institutional development is the effort to move corruption measurement beyond the ‘usual anti-corruption players’ such as ACAs and CSOs and into the national statistical system. The framework foresees national statistical offices, which already report progress on SDG 16.5, as having a crucial and impartial role in ensuring more integrated data collection with other public bodies such as ACAs, tax authorities, law enforcement agencies, audit bodies, and others that are already collecting relevant data. How such coordination and collaboration could work in practice has been tried in three countries.

Since 2024, UNODC, together with the UN Statistics Division's Data for Now initiative, has supported pilots adapting the framework to national needs and capacities in Colombia, the Dominican Republic, and Kenya, and developed a practical guide operationalise the UNODC statistical framework. In this interview, we asked three experts who have worked with the pilot countries and drafted the guide to share their experiences with these processes and emerging lessons learned for other countries embarking on more systematic collection of corruption data.

The experts we interviewed

We interviewed the following experts by e-mail in summer 2026.0f4b2ee22492

Byron Villacis Cruz is a Professor at the University of Oregon whose research examines how quantification, expertise, and power shape the production and use of corruption and anti-corruption knowledge. He is currently a consultant for the UN Statistical Division, where he works in collaboration with the UNODC with countries to operationalise the statistical framework and strengthen corruption-related statistics under the Data for Now initiative. Previously, he served as head of Ecuador’s National Institute of Statistics and Censuses, leading the 2010 Census and more than 35 national official surveys, and has advised statistical offices in Latin America and Africa.

Faryal Ahmed is a Statistician who leads the Data for Now initiative at the UN Statistics Division, driving efforts to strengthen national statistical systems and close critical data gaps across sectors such as environment, governance, crime, urban development, and food security.

Salomé Flores is the Head of the UNODC Information Centre for researching and analysing transnational threats related to drugs and crime, based in Tashkent, Uzbekistan.

Getting started: Translating ambition into action

The guide aims to help countries operationalise a large set of corruption indicators. What were the biggest challenges in making such a complex statistical framework usable for policymakers and practitioners?

One of the main challenges was bringing together institutions with different mandates, data systems, capacities, and policy priorities, and helping them agree on what should be measured, why it mattered, and who would be responsible for producing and using the statistics.

Countries were seeking stronger corruption statistics to guide policies that reflect national priorities, but their institutional arrangements and information needs varied considerably. Measuring corruption is not simply a technical exercise; it requires coordination among national institutes with varying capacities, agreement on shared priorities, and collective efforts to overcome bottlenecks to access data and produce the resulting statistics to inform policies.

The challenge was therefore to translate national demand into a process that was methodologically sound, institutionally feasible, and relevant to each country’s context. The UNODC statistical framework and the practical guide helped countries identify priorities, convene the relevant institutions, assess available data sources, and establish the statistical and coordination arrangements needed to produce credible and policy-relevant corruption statistics. At the same time, the guidance needed to remain flexible enough to accommodate different national contexts while preserving statistical quality and policy relevance.

Country experiences and implementation lessons

What were the needs and motivations that triggered implementation in Colombia, the Dominican Republic, and Kenya? Who was involved, what were the major milestones, what were the first outputs or outcomes, and what lessons were learned?

Several anti-corruption agencies and national statistical offices had contributed input to the UNODC Statistical Framework before its adoption and, through that process, expressed interest in receiving support to implement it. When UNODC and the UN Statistics Division began collaborating under the Data for Now initiative, the team engaged with several national institutions and identified clear interest and commitment in Colombia, the Dominican Republic, and Kenya.

In each country, an important first step was for the national statistical office and the anti-corruption agency to agree on how they would collaborate. Together, they began identifying corruption measurement priorities based on national strategies and policy needs, assessing the availability and quality of relevant data, and exploring which indicators from the statistical framework could serve as feasible entry points. As the process developed, they involved other institutions that produce or use relevant data, including audit and comptroller institutions, prosecution services, law enforcement and criminal justice bodies, financial intelligence and anti-money laundering agencies, procurement authorities, tax and customs administrations, and civil service institutions.

While the pilot processes were centered mainly on public institutions, the guide recognises an important complementary role for civil society and other non-governmental stakeholders in identifying blind spots, contributing independent evidence and contextual knowledge, and supporting interpretation, dissemination, and accountability.

The initial milestones and outputs included the establishment of coordination arrangements, the identification of priority measurement areas, the mapping of potential data sources, and the selection of an initial set of indicators for further development. These were not intended to constitute comprehensive national corruption measurement systems, but rather to create a practical foundation for longer-term work.

A key lesson is that countries should begin by identifying their most policy-relevant priorities and then sequence the statistical work realistically, taking institutional capacity and data availability into account. The indicators selected in the first phase should therefore be understood as entry points rather than as a complete representation of the national anti-corruption agenda. This reflects the flexible design of the statistical framework, which allows countries to focus on the areas and indicators most relevant to their own policy context and challenges.

Based on your experience in Colombia, the Dominican Republic, and Kenya, what made coordination between statistical offices and anti-corruption agencies actually work?

Coordination worked when the statistical office and the anti-corruption agency recognised their complementary roles. The statistical office contributes methodological credibility, quality assurance, documentation of standards, and a connection to the national statistical system and official statistics. The anti-corruption agency contributes policy relevance, knowledge of legal and institutional priorities, and convening capacity across the public integrity, justice, procurement, audit, and accountability ecosystems.

The relationship became stronger when both institutions saw the process as co-production rather than a simple request for data. Clear tasks also matter, including selecting priority areas, identifying data holders, assessing data quality, clarifying legal bases for data sharing, and defining how the resulting statistics will be used. Coordination is therefore not only an administrative requirement; it is integral to building a credible and useful national system of corruption-related statistics.

Consulting and agreeing on priorities

If a country wants quick but credible results, how should it identify its first three or four core indicators without overwhelming the system?

Quick but credible results come from sequencing, not shortcuts. A country can begin with a small core of indicators. However, the selection should be linked to national anti-corruption priorities, supported by plausible data sources, and owned by the institutions that will produce, validate, and use the information. The starting point should be the policy question: which corruption or anti-corruption issues does the country need to better understand at its current stage? Only then should feasibility and data availability be assessed.

It is important to note that the first indicators should be presented as entry points within a broader measurement agenda, not as a complete description of the country's anti-corruption priorities. This framing is important institutionally: selecting an initial set does not mean other dimensions are less relevant. It means the country is building a phased process, using a manageable first cycle to test coordination arrangements, improve data flows, demonstrate value, and create the conditions for expansion over time.

How did stakeholder consultations across governments, statistical offices, and civil society shape the design of the guide, and what tensions or trade-offs emerged?

Stakeholder participation should not be mechanical; it has to be organic. What gives a statistical system legitimacy is not only its technical quality but also its transparency and accountability and the fact that relevant actors can see a practical benefit in its existence, use, and sustainability. In this case, the consultations helped demonstrate that corruption statistics can be legitimised through consultation, trust-building, clarity of roles, and a shared understanding of how the information will be produced, validated, interpreted, and used. Engagement with government institutions, national statistical offices, anti-corruption bodies, civil society, and other users helped ensure that the guide addressed real implementation issues: how to convene the right institutions, clarify mandates, identify and validate data sources, manage data-sharing concerns, and communicate results responsibly.

The main trade-offs were between feasibility, ownership, and sensitivity. Countries could pursue broad or narrow anti-corruption agendas, but the first cycle of statistical work must be manageable, credible, and institutionally viable. Some key indicators may require improvements to administrative records, coordination mechanisms, or data quality before they can be produced responsibly. For that reason, the consultations reinforced an important principle: the framework should support nationally owned processes. Its value depends on helping countries move at a realistic pace while protecting institutional neutrality, statistical quality, and the trust of the actors involved.

Future directions and institutional capacity

How do you envision countries using the data generated through this framework to drive concrete anti-corruption policies or reforms?

The vision is to incorporate corruption statistics into the regular policy cycle. Indicators should not remain as isolated numbers in a report; they should help institutions identify risks, monitor reforms, understand bottlenecks, request disaggregated information, take actions, and assess whether preventive or enforcement measures are producing observable changes over time.

An important part of the process is identifying data sources that institutions already collect, particularly administrative data produced on a regular basis. Where these sources are sufficiently relevant and reliable, they can support the regular production of corruption-related indicators. This gives policymakers a more consistent view of trends over time and helps them assess whether risks are increasing, reforms are having an effect, or further action is needed.

In practical terms, the statistical framework can support a management approach to anti-corruption policy. Once institutions see that statistics help them prioritise, coordinate, and evaluate their work, they have a reason to request updates, improve administrative records, refine indicators, and strengthen collaboration with the statistical system. This creates a feedback loop between evidence and policy, which is essential for moving from general concern about corruption to more systematic and professional management of anti-corruption efforts.

Looking ahead, how do you see this guide evolving, especially in relation to global initiatives like the SDGs and the UNCAC implementation process?

The guide should evolve into a living tool that learns from implementation. As countries apply the statistical framework, adapt it to their contexts, and identify practical challenges, the guide can incorporate lessons on coordination, governance, validation, dissemination, and the responsible use of corruption-related statistics. Its purpose is not to freeze a single model but to support countries in building sustainable statistical capacity in a sensitive and policy-relevant area.

The link with UNCAC is especially important. Article 61 provides a clear basis for strengthening knowledge, analysis, and information on corruption, and data generated through the statistical framework can also support countries in reporting progress in the next phase of the UNCAC Implementation Review Mechanism. The connection with the SDGs is also relevant, particularly because two SDG indicators are included in the statistical framework. In this sense, the guide can help countries connect international commitments with national decision-making, ensuring that reporting obligations also strengthen domestic evidence and policy capacity.

If you could bestow some additional powers on statistics offices to do their work more effectively, what would they be?

Byron – Drawing on my experience in a national statistical office, I would strengthen professional independence: the authority to choose methods, standards, data sources, and publication timelines without undue interference. I would also give statistical offices a clearer government-wide coordination mandate, including reliable access to administrative data. Independence and coordination must go together. The first protects the credibility and impartiality of official statistics; the second allows statistical offices to respond to national priorities, reduce duplication, and improve the evidence available for policymaking.

Faryal – From my perspective, strong statistical legal mandate is essential because it ensures professional independence while giving national statistical offices the ability to collect data from various government entities and preserving confidentiality. That mandate should be matched by adequate and predictable resources and mechanisms that make inter-institutional cooperation a regular responsibility rather than an optional arrangement. These conditions would allow statistical offices to respond to emerging national priorities while safeguarding the quality and impartiality of official statistics.

Salomé – National statistical offices should have the power to ensure data quality through audits of administrative records and any other relevant data source. Audits will be useful to verify the accuracy, consistency, completeness, and uniqueness of data sources. Audit findings could inform remediation plans to improve data quality, facilitate harmonisation and support the use of analytics. This would make data easier to interpret and disseminate to a wider audience.

Disclaimer

The views expressed in this interview are personal and do not necessarily reflect those of the United Nations, its offices, officials, or Member States.