Saturday, November 12
Questionnaire Design
Sat, Nov 12, 9:00 AM - 10:25 AM
Hibiscus B
Adapting for the Web

Modernizing Data Collection of UK Labour Force Statistics for the 21st Century (303587)

Louise Morris, Office for National Statistics  
*Alex Nolan, Office for National Statistics  
Ian O'Sullivan, Office for National Statistics  
Laura Wilson, Office for National Statistics, UK 

Keywords: questionnaire design, new approaches, design methods, testing methods

The Office for National Statistics (ONS) recognizes the need to modernize our data collection strategies for the digital era, in particular implementing online self-completion modes for our Social and Business surveys, and the 2021 Census. We expect that providing an online offering as part of a redesigned, better integrated mixed-mode approach will help address declining response rates, while maintaining data quality and providing more timely access to statistics.

Taking the example of the Labour Force Survey (LFS), the UK’s principal measure for employment statistics, we will describe our online-first, mobile-first, and respondent-centered approach to mixed-mode design which necessitates significant changes to what is the UK’s largest household survey.

LFS content has grown substantially since 1973 as generations of policy makers have used the survey as a means of meeting new data requirements, while being reluctant to remove questions to preserve historical comparability. Moving this survey to online self-completion presents challenges to the way we collect data, redesigning a questionnaire and interviewer interaction that has worked well in the face-to-face interview to suit an online context. Through our work we aim to understand the shortcomings of existing questions and conduct respondent-based research around topic comprehension, online behaviors & usability and how these interact to affect respondent's processing of the question. Applying our framework of design principles for mixed-mode survey design, we take an iterative and multi-faceted approach to testing and redesigning our survey questions to be respondent-friendly, while meeting the data requirements. We aim to develop a more relevant and shorter respondent experience, while maintaining data quality.

Our development and testing methods to transform from a data-centered, to a respondent-centered approach to Questionnaire Design will be discussed, and findings presented.