BACK TO SCHEDULE SPEEDTALK

Speed talk session 5

Communicating Quality through Metadata

3 June 2026
13:15 – 14:00
ŠIBENIK V SHOW ON MAP

Session chair
Susana Portillo
Expert in Statistical Quality and Support Services, Central Statistics Office (CSO), Ireland

Read more Read less

Susana Portillo has worked in official statistics in the Central Statistics Office (CSO) in Ireland since 2007. From 2015 she has been involved in the design and delivery of the CSO quality strategy through the implementation of a solid Quality Management Framework. She currently leads the unit on Metadata and Quality Training, ensuring the harmonisation of questionnaire design and quality reporting across the CSO and their documentation conforming to international standards. She also provides support on metadata and quality techniques to the wider Irish National Statistical System. Prior to her work in quality, Susana led the unit responsible for the data collection of short-term enterprise statistics, using continuous improvement techniques to migrate statistical production to a process approach, achieving efficiencies in timeliness, response burden and cost while maintaining the quality of the production data. Susana is a graduate in Mathematics with a specialisation in Numerical Analysis, she holds a postgraduate diploma in Computer Science and is qualified in Lean Six Sigma techniques.

Presentation title
Strengthening Quality and Transparency Through AI-Ready Documentation: The SCAD Experience
PRESENTATION

National statistical offices are under growing pressure to communicate quality in ways that are transparent, consistent, and accessible to users.

Read more Read less

For the Statistics Centre Abu Dhabi (SCAD), this challenge has been central to its statistical modernisation agenda. Over the past year, SCAD has redesigned its documentation system to ensure that every statistical product is supported by clear and structured documentation. This paper presents our experience in developing AI-ready documentation, an approach that strengthens quality reporting while preparing official statistics for an increasingly automated future.

The transformation of documentation began with the need to move away from information scattered across Word, PDF, and Excel files toward a single, coherent system. The new documentation template is fully aligned with the GSBPM 5.2 sub-processes, implemented on a centralised Confluence platform, and supported by a clearly defined ownership model in line with the organisational operating model. Rather than treating documentation as an administrative requirement, we embed it within statistical production, capturing methods, metadata, quality indicators, and supporting evidence in a consistent structure across all statistical products.

A major milestone in this transformation was the documentation of the Household Income and Expenditure Survey conducted in 2024. This exercise demonstrated how end-to-end documentation enables the systematic measurement of statistical quality and strengthens the transparent communication of quality information across all phases of production, from survey design and data collection to processing, analysis, and dissemination. Evidence such as training materials, validation files, and dashboards is embedded within the documentation pages and mapped to the relevant GSBPM 5.2 sub-processes. This improves traceability, facilitates quality assessment, and supports capacity building by enabling staff to navigate statistical processes and understand how individual components contribute to overall quality.

The documentation project also aims to support the next generation of statistical production and dissemination. By ensuring that documentation and metadata are structured, standardised, and machine-readable, we have established a strong foundation for AI readiness. Artificial intelligence tools, notably SCAD’s Compass AI, can retrieve accurate and context-aware explanations of methods, metadata, and quality statements directly from the documentation. This strengthens institutional memory, supports staff capabilities, and improves internal communication around statistical processes and their quality.

Looking ahead, we are experimenting with AI-assisted documentation, where artificial intelligence supports drafting, updating, and maintaining documentation and metadata based on existing evidence and workflows. This will enhance how quality and metadata are communicated and reported to users, reinforcing transparency, trust, and the resilience of official statistics.

Main author / Presenter
Sapphire Yu Han
Statistics Centre Abu Dhabi (SCAD), United Arab Emirates

Read more Read less

Dr. Sapphire Yu Han achieved her PhD in demography at VU vrij university Amsterdam and she is a Statistical Quality Expert at Statistics Centre Abu Dhabi (SCAD). Her work focuses on statistical quality, official statistics, metadata, and AI-ready documentation. She is also Co-Lead of the AI-Ready Dissemination Project under the UNECE High-Level Group for the Modernisation of Official Statistics, supporting the development of approaches that enable official statistics to be more accessible and usable in AI-driven environments. She previously worked with UN, OECD, and Statistics Netherlands.

CO-AUTHORS:

Presentation title
Measuring and Disseminating Quality Indicators on Statistical Outputs: The Standards and Tools Adopted by FAO
PRESENTATION

Transparency and trust in official statistics require systematic measurement and clear communication of data quality.

Read more Read less

The Food and Agriculture Organization of the United Nations (FAO) evaluates the quality of its statistical outputs across five dimensions, as defined in the FAO Statistics and Data Quality Assurance Framework (SDQAF): relevance, accuracy and reliability, timeliness and punctuality, coherence and comparability, and accessibility and clarity.

This paper outlines the methodologies and IT tools adopted to measure these dimensions—particularly for secondary data, which FAO primarily compiles from national authorities and other international organizations—and to disseminate quality indicators that inform external users about the strengths and limitations of FAO statistics. Measuring quality for secondary data presents methodological challenges, as direct accuracy measurement is often not feasible. FAO relies on a combination of quantitative indicators (e.g., completeness, timeliness, imputation rates) and qualitative assessments (e.g., metadata on sources, revision policies, and user feedback) to evaluate reliability and transparency. This dual approach enables FAO to assess not only the statistical outputs but also the processes underpinning their production, reinforcing accountability, and trust in global food and agriculture statistics.

Dissemination of quality indicators is integral to FAO’s commitment to clarity and accessibility. Indicators are systematically published alongside statistical outputs through standardized reference metadata templates, which represent the preferred channel for communicating quality information in a comparable and structured way. The standardized computation of these quality indicators not only ensures consistency across domains but also facilitates the compilation of reference metadata.

Beyond external communication, quality indicators serve as a critical internal tool for monitoring and continuous improvement. They help identify gaps and challenges in data quality, inefficiencies in data collection processes, and areas requiring methodological enhancements. These insights inform FAO’s work planning, resource allocation, and engagement strategies with national authorities, ultimately strengthening the statistical production process.

Main author / Presenter
Francesca Rosa
Food and Agriculture Organization of the United Nations (FAO), Italy

Read more Read less

Francesca currently works as Statistician in the FAO Data Quality Unit, where she contributes to the development and update of statistical standards, the implementation of quality assessments and peer reviews, and supports technical units in operationalizing statistical standards. She previously worked across several international organizations, including FAO, UNODC, and WFP, gaining extensive experience across all phases of the data lifecycle—from collection and validation to analysis, visualization, and dissemination.

Presentation title
The GeoSTAT Metadata Catalogue – Enhancing Metadata Access, Integration and Use
PRESENTATION

The GeoSTAT Metadata Catalogue, developed by the Croatian Bureau of Statistics (DZS), is an open metadata repository implemented on GeoNetwork catalog application designed to support discovery, access and integrated use of spatially-referenced data and information across multiple statistical and geospatial domains.

Read more Read less

It aims to address key challenges in spatial data sharing by providing a centralized, searchable metadata platform that facilitates the discovery of datasets and services produced within the national statistical system and disseminated on the GeoSTAT portal. The catalogue enhances decision-making processes, multidisciplinary integration and understanding of geographic information benefits for both producers and users of official statistics. By adopting internationally recognized metadata standards, the catalogue improves the visibility and usability of metadata records, thus improving interoperability between providers and users of spatial data and its services. The catalogue supports structured metadata discovery, including title, summary, keywords, geographic coverage, temporal extent, lineage, compliance and access links. Its functionalities include simple and advanced search, metadata preview, full metadata view and metadata export in multiple formats, thus fostering broader accessibility for researchers, policymakers and data users. This presentation will outline the role of metadata catalogues in strengthening official statistics ecosystems in line with quality frameworks and user needs. Overall, the GeoSTAT Metadata Catalogue exemplifies how spatial and thematic metadata enhance quality, interoperability and user engagement.

Main author / Presenter
Branko Crkvenčić
Croatian Bureau of Statistics (CBS)

Read more Read less

Professor of geography, working in Croatian Bureau of Statistics from 2005. Head of Geoinformation system development unit. Work experience with merging statistics and geoinformation information in this areas: Merging of statistical data of the 2011 Census of population, households and dwellings with geospatial information in the Republic of Croatia, Merging statistics with geospatial data in member states - business register data, Merging data on accommodation establishments and tourist activity with geospatial information in the Republic of Croatia.

CO-AUTHOR: