Designing and delivering online surveys: Transitioning long questionnaires
Date:
14/07/2021
Organised by:
GenPopWeb2, University of Southampton
Presenter:
Ian O'Sullivan, Tim Hanson and Peter Lugtig
Level:
Entry (no or almost no prior knowledge)
Contact:
Dr Olga Maslovskaya
om206@soton.ac.uk
Venue: Online
Description:
GenPopWeb2 is a network of UK-based academic and non-academic partners to share knowledge and collaborate in the area of online data collection in social surveys as well as in setting the research agenda in the area. It is funded by the Economic and Social Research Council (ESRC).
We are pleased to present two online workshops (via Zoom) on the theme of designing and delivering online surveys. Each of the events will focus on different aspects of online survey design and implementation; the first will focus on transitioning long questionnaires to online and the second will focus designing, developing, and testing online questionnaires. The second event will take place on the 30th of July and will be announced shortly.
Each event will consist of three invited presentations followed by an extended discussion. The aim is to showcase current evidence, experiences, and approaches. Discussion will focus on advice, challenges, and opportunities.
Wednesday 14th July 2021, 15:00-16:45 (UK time): Transitioning long questionnaires to online
15:00-15:05
Introduction
15:05-15:25
Transforming social surveys in ONS: opportunities and challenges
Ian O’Sullivan (Office for National Statistics, UK)
ONS has been progressing a programme of work over the last four years to transform both social business and data collection. This has included building the capability to introduce a default online mode of collection for the Labour Force Survey. Initial discovery qualitative work demonstrated that ‘simply’ lifting and shifting the LFS content and including it on a self-completion online questionnaire would fail so a decision was taken to fully transform the LFS content. This approach adopted some key principles two of which were to reduce the size of the questionnaire and to put the respondent at the centre of the redesign. This approach has shaped the research and design journey that ONS has taken in recent years. Over this period ONS has also been working towards moving towards the development of an integrated statistical framework for all social surveys into a single integrated statistical framework. This model aims to modularise all ONS’ social surveys around a large-scale population survey which would also support ONS’s strategic ambition to use administrative data sources to produce population and social statics between (and potentially in place of) decennial Censuses. This talk provides an overview of what we have learnt about moving the LFS online and further details on where we are with the integrated survey design.
15:25-15:45
Developing self-completion instruments for the European Social Survey
Tim Hanson (City, University of London, UK)
This talk will give an overview of development and testing to produce self-completion (web and paper) instruments for the European Social Survey (ESS). The self-completion method has been developed for Round 10 of ESS as a back-up approach for countries that are unable to deliver the usual face-to-face approach due to the pandemic. We will look at some of the challenges of converting an hour-long face-to-face survey to self-completion and share results from user testing and experimental work to assess effectiveness of the instruments.
15:45-16:05
Moving a long survey online. Problems and some potential solutions
Peter Lugtig (Utrecht University, Netherlands)
When face-to-face or telephone surveys are moving online, they are ideally redesigned and reconceptualized to fit the online environment. One key issue here is survey length. This talk will present empirical evidence from studies which have documented what happened when long or very long surveys were moved online without redesign. Potential risks here are that respondents may be less keen to start a long online survey, that respondents may break-off, may give worse answers, or decide not to participate in future surveys after participating in a long survey. How much evidence is there for such effects? One potential solution to move long surveys online is by modularizing it. Instead of fielding one long survey, several shorter surveys are fielded. Over the past years, several studies have experimented with modularization with mixed success. The presentation will present some of these findings and discuss whether modularization is a potential route for moving long surveys online.
16:05-16:35
Discussion
16:35-16:45
Wrap-up and close
Cost:
This workshop is free to attend but please register via the link below.
Website and registration:
https://us02web.zoom.us/meeting/register/tZMlc-iprTMpGdMYDUu5VO8tBsKKZuRArJfw
Region:
South West
Keywords:
Online Data Collection
Related publications and presentations: