by Clara Aida Khalil, Team Leader of the Data Quality Unit of FAO Statistics Division.

In today’s world, data and statistics are the backbone of decision-making, guiding the design of sound policies and targeted interventions, and enabling governments and the international community to act on evidence rather than assumptions. Therefore, considering that decisions are only as strong as the information behind them, producing high-quality statistics is an essential duty of national and international statistical organizations.

The Food and Agriculture Organization of the United Nations (FAO) has carried this responsibility since its creation in 1945. Article I of its Constitution mandates the Organization to collect, analyse, interpret and disseminate information on nutrition, food and agriculture, making statistics a core pillar of FAO’s mission. Today, as custodian of 22 Sustainable Development Goal (SDG) indicators and producer of authoritative data on agriculture, forestry, fisheries, food security and beyond, FAO is both a global data provider and a standard-setter, striving to ensure that its statistics are relevant, reliable, timely and accessible.

The challenge of ensuring quality in a complex statistical ecosystem

FAO operates in a uniquely complex statistical environment. Its decentralized statistical system brings together the Statistics Division and many technical units producing thematic data across FAO’s mandate. Coordinating this system requires strong governance and robust quality assurance mechanisms to ensure that every output meets the same high-quality standards.

This challenge is compounded by the fact that most FAO statistics rely on secondary data provided by countries. In these cases, national statistical agencies are the primary producers of official statistics, while FAO compiles, validates and harmonizes their outputs to ensure global comparability and accessibility. The reliability of FAO statistics therefore depends directly on the quality of national data. To reinforce this, FAO works in partnership with national counterparts—providing capacity development and technical assistance, sharing methodologies and promoting international standards—to strengthen national statistical systems and support the production of high-quality, internationally comparable data.

Adding to this complexity is the rapidly evolving data landscape. The rise of big data, geospatial information and other non-traditional sources offers opportunities to improve timeliness, detail and insight. However, it also raises concerns around privacy, reliability and sustainability. FAO must therefore strike a careful balance, embracing innovation while safeguarding statistical integrity, so that new data sources enhance rather than compromise quality.

A long and steady journey towards effective governance and quality assurance

FAO’s commitment to statistical quality is not new. In 2014, the Organization adopted its first Statistics Quality Assurance Framework (SQAF), including a series of quality principles to adhere to accompanied by good practices for their implementation. These principles reflect the Fundamental Principles of Official Statistics as well as principles endorsed by the Committee for the Coordination of Statistical Activities (CCSA).

Nearly a decade later, in 2023, the SQAF was revised and expanded into the Statistics and Data Quality Assurance Framework (SDQAF). The update responded to profound shifts in the data ecosystem, including the growing use of non-traditional data sources and concerns around privacy and intellectual property. The SDQAF extends FAO’s quality principles to these new domains, ensuring that innovation strengthens rather than undermines trust in official statistics.

The framework is built on 16 core principles providing a comprehensive blueprint for producing high-quality statistics that are fit for purpose. They cover three key areas:

• The institutional environment - including professional independence, commitment to quality, data protection, use of international standards, and cooperation with data providers and international organizations.
• The statistical processes - ensuring the use of trustworthy data sources, sound and cost-effective methodologies, and limited burden on data providers.
• The statistical and data outputs - which must be relevant, accurate and reliable, timely and punctual, coherent and comparable, clear and accessible.

The SDQAF is reinforced by a strong governance system. Since 2012, the Chief Statistician has overseen the coordination of FAO’s decentralized statistical system, ensuring coherence and consistency. The Data Coordination Group (DCG)—chaired by FAO’s Chief Economist as Executive Data Champion—and its technical arm, the DCG-T, chaired by the Chief Statistician, set priorities, address cross-cutting issues and develop FAO’s statistical standards.

In 2024, FAO further strengthened this architecture with a dedicated Data Quality Unit in the Statistics Division. The unit maintains and updates the SDQAF and statistical standards, conducts quality assessments and peer reviews, coordinates user consultations on FAO databases, and delivers internal and external training and assistance on quality-related matters. Its role is to embed a culture of quality across the Organization, ensuring that every statistical product produced or disseminated by FAO meets the highest standards.

The FAO’s SDQAF in action: the quality assurance mechanism

The SDQAF is implemented through a well-structured quality assurance mechanism that operates across all of FAO’s statistical activities. This mechanism combines the endorsement of a set of statistical standards and best practices, quality assessments, user consultations and administrative clearances to ensure that every data product meets the highest quality standards.

As of now, the Organization has developed 18 statistical standards providing general and technical recommendations and clarifying the governance procedures for key activities along the entire statistical production process (i.e. for data collection, processing, documentation, and dissemination) as well as for specific cross-cutting quality issues.

The implementation of the SDQAF and the related statistical standards is evaluated by means of various kinds of quality assessments. These include self-assessments, in-depth reviews, standards-based evaluations and peer reviews. They are carried out by the Data Quality Unit, at times with the collaboration of internal or external subject-matter experts. These assessments are aimed at producing a set of actionable recommendations and improvement plans with clearly defined responsibilities and expected outcomes.

Equally important are user consultations, capturing the views of users of FAO’s statistics on the relevance and quality of the statistical and data outputs produced by the Organization. As of now, most of the largest FAO statistical databases underwent at least one user consultation, that resulted in a better understanding of the characteristics and needs of FAO’s users and in the identification of areas for potential improvement.

Finally, statistical and data-related clearances provide a safeguard before new initiatives are launched. As part of FAO’s accountability framework for statistical activities, all new data collections, statistical projects and initiatives and dissemination platforms undergo a review and clearance process to verify methodological soundness, cost-effectiveness and compliance with quality standards.

Together, these mechanisms translate the principles of the SDQAF into practice, ensuring that FAO’s statistics remain trusted, relevant and of consistently high quality.

High-quality data for an inclusive future

FAO’s statistical system continues to evolve, adapting to the opportunities and challenges brought by new technologies and diverse data sources, while upholding its core mandate to produce high-quality statistics on food and agriculture.

By embedding quality assurance and management at the heart of its operations, the Organization is well positioned to harness innovation responsibly, ensuring that its statistical outputs remain relevant, reliable and useful for years to come. As FAO marks its 80th anniversary, and as the world celebrates the World Statistics Day, it is therefore fundamental to reaffirm the commitment to produce high-quality data and statistics to inform decisions and help shape a food-secure future for all.

Ms Clara Aida Khalil is the Team Leader of the Data Quality Unit of FAO’s Statistics Division. In her role, she coordinates the development and revision of FAO Statistical Standards, and the implementation of quality assessments on FAO statistical and data processes and outputs.