Newsletter

📌 In this Newsletter:

• What we do

• How to join

• Invited session at IDWSDS

• Coming up: SIG webinar - Vera Toepoel

Dear colleagues,

Welcome to the first newsletter of the Multisource Statistics Special Interest Group of the European Survey Research Association (ESRA).

What we do

As survey research increasingly operates in a complex data ecosystem, integrating probability-based surveys with administrative records, digital traces,  non-probability samples, and other non-traditional data sources has become essential for improving inference and for studying phenomena that cannot be adequately measured by a single data source. The MSS-SIG provides a forum for exchange, collaboration, and methodological advancement within ESRA’s core mission.

Our topics include, but are not limited to:

• Integration of surveys, administrative data, registers and digital trace data

• Inference using multiple sources (bias, variance, uncertainty propagation, small area estimation)

• Inference with non-probability samples (calibration, weighting, model-based approaches)

• Integrating surveys and digital traces (web data, social media, sensors, paradata)

• Record linkage and statistical matching

• Quality assessment and validation in multi-source settings

• Ethical, legal, privacy and governance challenges
  
  

How to join

To join the group and subscribe to this newsletter, fill in the short form on our website: Join the SIG. Please feel free to forward this to colleagues who might be interested.

Website: https://sites.google.com/view/sig-multisource-statistics/home

Share your ideas with us

This SIG is meant to be shaped by its members. Do you have ideas for topics, webinars, workshops, collaborations or special sessions? Let us know through this form or write to us directly.

Where we’ve been: invited session at IDWSDS

We organized the invited session “Advances in Multisource Statistics” at the International Day of Women in Statistics and Data Science (IDWSDS) conference, on 6 October. Our speakers, Monica Pratesi (University of Pisa), Char Hilgers (DIW Berlin), Jisu Kim (Utrecht University) and Manuela Schmidt (RPTU Kaiserslautern-Landau), brought different perspectives on multisource statistics and their impact on society. The session abstracts are available here.

Coming up: SIG webinar by Vera Toepoel

Date and time: 25th Nov, 11-12

Registration: here

Title: Beyond Traditonal Surveys and Towards Innovative and Integrated Multi-Source Statistics

Speaker: Vera Toepoel, Statistics Netherlands

Bio:

Vera Toepoel is head of the methodology department at Statistics Netherlands in Heerlen. Since May, she is Professor on the topic of data collection at the department of Methods and Statistics at Utrecht University, where she worked from 2012-2022. Her research focuses on online survey design, mobile data collection, and innovative digital research methods. She has authored textbooks, including "Doing Surveys Online" by SAGE. Additionally, she had global leadership roles, serving as the President of the European Survey Research Association and the President of RC33 for the International Sociological Association.

Abstract:

Official statistics face a changing data landscape and growing expectations from society. The ambition is to become more innovative, integrated and efficient, while reducing respondent burden and safeguarding statistical quality and relevance. Achieving this requires making better use of the growing range of available data, including administrative records, publicly available data, privately held data and information collected through traditional surveys. Combining these sources offers important opportunities for developing high-quality, relevant and timely statistics, but also raises fundamental methodological challenges.

A central question is how different data sources can be integrated while maintaining representativeness, measurement quality, consistency and public trust. New sources may offer richer information and reduce reliance on conventional questionnaires, but they can also introduce coverage biases, selective participation, comparability issues and new demands for data governance. Innovation therefore requires more than access to additional data: it requires robust methodological frameworks for evaluating sources, understanding their limitations and determining how they can complement one another.

Rather than viewing new data sources as replacements for traditional surveys, the presentation advocates a complementary approach in which different sources are evaluated and combined according to their strengths and limitations. The ultimate goal is to develop robust, transparent, and responsible methods that make better use of available data while preserving the reliability of official statistics.

The presentation will conclude by inviting discussion on the methodological priorities and collaborative research needed to advance multisource statistics across national statistical institutes and the wider survey research community.

Camilla Salvatore and Angelo Moretti

Chairs - ESRA Multisource Statistics SIG