Date/Time
Date(s) - 24/11/2025
3:00 pm - 4:30 pm
Category(ies) No Categories
IASS Webinar 58: R-indicators for Assessing Representativeness for Surveys, Administrative Data and Non-probability Samples
Speaker: Natalie Shlomo
University of Manchester
Monday, November 24th 2025 at 3pm – 4:30pm (CET)
All are invited to the webinar, organised by the International Association of Survey Statisticians.
Please register for the IASS Webinar at:
https://us06web.zoom.us/webinar/register/9417625059240/WN_GL33LRqLR7K0qVIO7hkJiA#/registration
After registering, you will receive a confirmation email containing information about joining the webinar. There will be time for questions. The webinar will be recorded and made available on the IASS and ISI web site. See below for the abstract and biography of the speakers.
Webinar Abstract
In this presentation, we focus on the use of Representativity (R-) Indicators to assess representativeness in survey and non-survey data. We start with the initial purpose of developing R-indicators for use in adaptive survey designs in our probability-based surveys. The R-indicators measure the degree to which respondents and non-respondents differ from each other (the contrast) according to a set of covariates available for both respondents and nonrespondents. R-indicators go beyond response rates that are traditionally used to monitor data collection. Through the analysis of R-indicators, we can build profiles (characteristics) of the data units where more or less attention is required in the data collection. We present an application of an adaptive survey design using R-indicators according to a structured trial and error approach and demonstrate the effectiveness of targeted data collection in a Dutch Crime Victimisation Survey.
R-indicators were extended to allow for assessing representativeness in survey data when information about the nonrespondents is not available. For this purpose, we can use population marginal and cross-tabulations to use as ‘plug-ins’ under a linear regression for estimating response propensities and R-indicators. We demonstrate with an application where R-indicators were successfully applied to assess representativeness in the 2011 EU-SILC datasets using 2011 European census counts as population benchmarks.
More recently, R-indicators have been expanded to assess representativeness in non-survey data sources, namely administrative data and non-probability samples. We can use population benchmarks when they are available for the case of a non-probability data source. However, when assessing representativeness on a continuous basis, we propose that statistical agencies make use of their high-quality survey data collections for population benchmarks. We show that we can use detailed weighted survey benchmarks to assess representativeness in administrative data with little addition to the variance of the R-indicators.
Biography
Natalie Shlomo is Professor Emerita of Social Statistics at the University of Manchester. Her main areas of interest are in topics related to survey statistics and survey methodology. She has over 80 publications and refereed book chapters and a track record of generating external funding for her research. She is an elected member of the International Statistical Institute (ISI), a fellow of the Royal Statistical Society, a fellow of the Academy of Social Sciences and Past-President 2023-2025 of the International Association of Survey Statisticians. She also serves on national and international Methodology Advisory Committees and editorial boards. Homepage: https://www.research.manchester.ac.uk/portal/natalie.shlomo.html