BACK TO SCHEDULE SPEEDTALK

Speed talk session 9

Inside the Quality Toolbox

4 June 2026
13:15 – 14:00
ŠIBENIK V

Presentation title
Applying the Total Error Model in Quality Reviews – A Practical Approach
Quality reviews are a key instrument for assessing compliance with the quality principles defined in the European Statistics Code of Practice.

Read more Read less

Their purpose is to identify areas to strengthen adherence to the principles. Since 2022, Statistics Norway has introduced the Total Error (TE) Model within its quality reviews. This paper presents a practical approach to applying the TE Model in reviews. This approach aims to uncover uncertainty and potential sources of errors in data, with a particular emphasis on CoP Principle 12: Accuracy and Reliability.

Methods:

The approach builds on Zhang’s Total Error Model (2012), which provides a structured conceptual framework for identifying and decomposing error sources within a data lifecycle. The approach is operationalized through a set of guiding questions addressing two dimensions:

- Representation (objects): Frame error, Selection error, and Missing redundancy. Examples include: Which objects should ideally be included in the source? Are there missing or overrepresented objects?

- Measurement (variables): Validity error, Measurement error, and Processing error. Examples include: What is the ideal information (target concept) versus what is actually measured? What is the difference between the intended measure and the obtained values? What are the sources for processing or editing errors?

These questions form the basis of a structured workshop including the quality team and the subject matter experts responsible for the statistics. The dialogue is focused on identifying possible sources of error and uncertainty, not specifically on the figures. The output – a list of potential error sources and uncertainties – is one of several inputs used to develop recommendations for improvement actions. These recommendations are discussed with the statisticians, and agreed actions are followed up regularly by the quality team.

Results:

Experience shows that TE Model- workshops effectively identify error sources and uncertainty early in the production process. Participants report increased awareness of uncertainty and the importance of measuring and communicating it. This method complements other review activities such as self-assessments, process analysis, and user focus groups, which together provide a comprehensive basis for identifying improvement areas.

Conclusions:

By operationalizing this theoretical model through practical workshops, Statistics Norway bridges the gap between conceptual error frameworks and actionable quality improvements. Integrating the TE Model into quality reviews provides a systematic and practical tool for improving statistical processes and outputs. It supports the Q2026 theme Enhancing Trust, Shaping the Future by promoting transparency and accountability.

Main author / Presenter
Morten Qvenild Andersen
Statistics Norway

Read more Read less

Morten Qvenild Andersen is a Senior Advisor at Statistics Norway and has been part of the Quality Team since 2025. Mr. Andersen began his career at Statistics Norway in 1999 and has extensive experience in statistical production, project management, and leadership in developing solutions for data editing. He is also a specialist in Agile methodologies, drawing on his long-standing experience as a Product Owner and Scrum Master, including serving as Product Owner for common data editing solutions at Statistics Norway. Mr. Andersen has worked on statistical quality for several years and has significant expertise in metadata, classifications, and standards. He is currently a member of the Secretariat for the Standards Committee at Statistics Norway and has served as a committee member for more than a decade.

CO-AUTHOR:

Presentation title
Internal Audits: An Effective Quality Control Tool
Internal audit is a key component of the Quality Management System (QMS) in statistical organizations, ensuring that processes comply with established standards such as ISO 9001 and that continuous improvement is achieved.

Read more Read less

Internal audits are the foundation of effective quality control tools. The primary goal of internal audits is to assess the effectiveness, consistency, and compliance of statistical processes and outputs, including data collection, processing, analysis, and dissemination.

In a statistical organization, internal audits evaluate whether operations adhere to documented procedures, meet user needs, and align with principles of official statistics such as relevance, accuracy, timeliness, and confidentiality. Audits also help identify non-conformities, root causes, and opportunities for improvement.

The internal audit process typically involves planning, preparation, fieldwork, reporting, and follow-up. Trained auditors - often internal staff independent of the process being audited -examine documentation, interview personnel, and observe procedures. Findings are recorded in audit reports and lead to corrective actions.

Incorporating internal audits strengthens transparency, enhances data quality, and builds public trust in statistics. It also supports the implementation of international frameworks such as the Generic Statistical Business Process Model (GSBPM) and Code of Practice for Official Statistics.

Main author / Presenter
Vugar Mammadalizade
State Statistical Committee

Read more Read less

Vugar Mammadalizade: He was born in 1979. He graduated from the Azerbaijan State University of Economics with a bachelor's degree in 1999 and a master's degree in 2008. In 2010, he graduated with honors from the Academy of Public Administration under the President of the Republic of Azerbaijan, specializing in "Public and Municipal Administration. From 2002 to 2015, he worked at the State Statistical Committee as an economist, senior consultant, sector head, and deputy head of department. Since 2016, he has been serving as the head of the Quality Management and Metadata Department of the State Statistical Committee. He has participated in numerous international seminars and conferences.

CO-AUTHOR:

Presentation title
Quality as an Inspiration for Standardisation
A culture of quality is a core element of the European Statistical System (ESS) and a key prerequisite for effective and sustainable standardisation within statistical institutions.

Read more Read less

In line with the ESS commitment to quality, quality is understood as a shared responsibility across all organisational levels and statistical domains. Beyond compliance and control, quality serves as a strategic framework that supports sound governance, harmonised processes, and the consistent production of high-quality statistical outputs.

This paper explores quality as an enabler of standardisation through two interrelated dimensions. The first concerns institutional quality arrangements, including leadership commitment, clear quality policies and objectives, and governance structures aligned with the European Statistics Code of Practice. The second focuses on operational quality, addressing standardised statistical processes and outputs structured in accordance with the Generic Statistical Business Process Model (GSBPM). Systematic process documentation, regular reviews, quality reporting, internal audits, and corrective and preventive actions ensure transparency, stability, and continuous improvement throughout the statistical production cycle. Decision-making based on data and an understanding of variation is central to producing robust and reliable statistics.

The paper demonstrates that a strong culture of quality delivers clear benefits for the ESS. Quality-driven standardisation enhances user satisfaction by ensuring relevance, accuracy, coherence, and timeliness, thereby strengthening trust and credibility. It also improves operational efficiency through process optimisation, error reduction, and better use of resources, while fostering staff engagement, competence development, and active participation in quality improvement.

Central to the ESS quality framework is the human dimension, recognising both producers and users of statistics as key stakeholders. Standardisation is therefore not an end in itself, but a means to ensure accessible, comparable, relevant, and trustworthy statistics. In an environment of increasing data complexity and information overload, a strong culture of quality remains essential for safeguarding public trust and ensuring the long-term sustainability of official statistics within the ESS.

Main author / Presenter
Ljiljana Okuka
Agency for Statistics of Bosnia and Herzegovina

Read more Read less

Ljiljana Okuka is a Senior Expert Advisor at the Agency for Statistics of Bosnia and Herzegovina, where she has been working for 16 years in the areas of statistical quality, methodologies, and standardisation within the European Statistical System. She holds a Master of Science degree in Business Finance and Banking from the Faculty of Economics in Pale, with a thesis on labour taxation and employment policies in Southeast Europe. She is actively involved in developing and implementing quality standards for statistical surveys and quality management frameworks. A certified trainer for Quality Management in Public Institutions under the CAF model, she also completed a professional traineeship at the Hellenic Statistical Authority (ELSTAT) upon recommendation by Eurostat. She is fluent in English, Greek, and Russian.

Presentation title
Quality Management during a challenging period
Over recent years, the UK Office for National Statistics (ONS) has experienced a period of substantial challenge, marked by some high-profile errors and heightened scrutiny of specific statistical outputs.

Read more Read less

These events, alongside reviews and enquiries from Parliamentary Committees, placed considerable pressure on staff and processes at a time when trust in official statistics was under close examination. This presentation explores how the ONS’s Quality Management framework played a central role in navigating this difficult period and supporting organisational recovery.

The session will outline how a flexible, innovative, robust and structured approach to quality—encompassing quality culture, risk management, transparent communication with users, and constructive engagement with the regulator—enabled the ONS to identify vulnerabilities, implement targeted mitigations, and drive continuous improvement. By reflecting on lessons learned and demonstrating how quality principles were operationalised in practice, the presentation aims to highlight the value of an embedded quality framework in strengthening resilience, supporting transparency on issues and challenges, and sustaining confidence in official statistics.

Main author / Presenter
Rachel Skentelbery
ONS

Read more Read less

Rachel Skentelbery leads the Quality and Improvement Division at the Office for National Statistics (ONS), and is the Deputy Head of Profession for Statistics for ONS. Rachel’s team provide support within ONS and across government on statistical methods and quality improvement. The division is focused on continuous improvement through the implementation of automation (through Reproducible Analytical Pipelines and Statistical Methods Library), best practice modelling, and implementation and understanding and improving quality. Previously Rachel worked at the Office for Standards in Education (Ofsted) as their Head of Statistical Quality and Data Science. Rachel is passionate about improving the quality, efficiency and usefulness of statistical products through the sharing of best practice, working in partnerships and the utilisation of new data and tools.