Tuesday 16th July, 09:00 - 10:30, D15 - Conference opening and Keynote Address 1
"Hybrid data collection in official statistics: Modes, devices and sensor data"
Prof. Dr. Barry Schouten (Statistics Netherlands and Utrecht University)
Today, mobile devices and wearables are standard tools for communication and personal archiving. Their rapid adoption but, simultaneously, their diversity and continuous change in shape and features, present both a great potential and a great challenge for primary data collection. The various devices closely follow a much larger trend, that of online communication and data collection. Over the last decade, official statistics introduced the online survey mode at large scale in surveys, mostly as a first mode preceding more expensive modes. The introduction has presented and still presents both methodological and practical challenges. Comparability over time and between subpopulations has become an issue to be taken very seriously. From a logistics point of view, combinations of modes are much more complex to manage and monitor. Mobile devices and wearables further challenge, and enrich, methodology and implementation. Mobile devices and wearables also bring sensors and sensors allow for automated measurements. These sensor data may be initiated by inviting respondents to enable the sensors in studies. In addition, these sensor data may exist already and may be linked after respondent consent. In the presentation, the migration to online surveys and surveys with devices and sensors is described following a number of real survey examples. A set of criteria is discussed for judging whether hybrid forms of data collection are promising and sensible and how they affect methodology and data collection practice.
Barry Schouten joined the methodology department of Statistics Netherlands in 2002 after a PhD in Mathematical Statistics at VU University Amsterdam. He worked on research and development projects in nonresponse adjustment, nonresponse reduction, mixed-mode survey design, mode-effect adjustment and adaptive survey design. Het published various papers on these topics. In 2009, he became a senior methodologist and coordinator of mixed-mode survey methodology at Statistics Netherlands. In 2017 he was appointed as professor at Utrecht University with a chair on the same topic. Prior to this, in 2016, Statistics Netherlands and Utrecht University established the joint research program on ICT innovation in data collection, called WIN, which he coordinates with Peter Lugtig. Within this program, research is launched into smartphone surveys and surveys with sensors.
Thursday 18th July, 11:00 - 12:30, D15 - Keynote Address 2 and Panel Discussion
"Best practices in survey research: are they really good and why don’t we implement them?"
Dr. Ineke Stoop, The Netherlands Institute for Social Research, SCP
There exists a wide range of literature on total survey error, best practices in survey research, survey guidelines and quality criteria for survey studies. There are gold standards, scientific handbooks on survey methodology, and codes of conducts showing us what to do. The industry, governments and academia each in their own way pursue survey quality. Survey researchers share values and assumptions about what makes a survey a good survey. The presentation will discuss a number of these assumptions by trying to answer the following questions:
- Are they correct?
- If they are true, why don’t we always implement the good practices?
- If they are sometimes or possibly true, when and under which conditions?
- If they are not true, what is the alternative?
A kaleidoscopic view of what we (maybe) know and don’t know (yet) about survey research will show why we will always need good surveys.
The presentation will be followed by a panel discussion.
Ineke Stoop retires from her job on the 14th of July 2019. Until then she was Head of Methodology at The Netherlands Institute for Social Research¦SCP, a government agency which conducts research into the social aspects of all areas of government policy. She also was Deputy Director Methodological of the European Social Survey and Chair of the European Statistical Advisory Committee. Dr Stoop obtained her Ph.D. in social sciences from Utrecht University for a thesis on survey nonresponse. She has taught courses on nonresponse, survey design and comparative surveys, and has authored articles, chapters and books on the same topics. In 2018 she was co-editor of a Wiley monograph titled Advances on Comparative Survey Methodology. In 2019 she will draft a chapter on nonresponse for the SAGE Encyclopedia of Research Methods. Ineke attended all previous seven ESRA conferences.
Panel Discussion Chair - Gerry Nicolaas
Gerry Nicolaas is Director of Methods at the National Centre for Social Research (NatCen) in the UK. She has worked as an applied Survey Methodologist for almost 30 years in government, academia and the non-profit sector. Gerry works at the intersection of science and practice to improve the quality of survey data collection. Her areas of expertise include survey response rates, respondent incentives, mixed-mode surveys, mode effects, web surveys and survey paradata. Throughout her career, she has worked on the methodology of many high-profile surveys in the UK, including the Health Survey for England, the UK Longitudinal Household Survey, the National Travel Survey, the Longitudinal Study of Young People in England, and the General Practice Patient Survey. In the last five years, her attention has shifted to the development of push-to-web surveys in the UK, focusing on respondent communications, device-agnostic questionnaire design, and transitioning offline to online data collection.
Annette Jäckle is Professor of Survey Methodology at the Institute for Social and Economic Research at the University of Essex. She is an Associate Director for the Understanding Society household panel study and leads a programme of research on innovations in data collection methods. Her current research focuses on developing best practice in producing integrated data: combining questionnaire based survey data, with measurement using new instruments such as apps, sensors, or wearables, and linked data.
Dr. Leyla Mohadjer is a Vice President and an Associate Director of the Statistical Staff at Westat, USA. She is the director of sampling and other statistical activities for several major international and national surveys, including the Program for the International Assessment of Adult Competencies (PIAAC), the Programme for the International Student Assessment (PISA) for Development, and the U.S. National Health and Nutrition Examination Survey. Her special areas of expertise include sample design and estimation in complex surveys, and development of standardized procedures and quality control guidelines following the Total Survey Error (TSE) framework, especially in multi-national, multi-cultural, and multi-regional studies. She has conducted research on topics such as the impact of incentives on the quality of data collected in surveys, nonresponse bias analysis, sampling of rare populations within the context of national surveys, and compositing of samples selected from overlapping frames. She is a member of the Westat Fellows Committee, an elected member of the International Statistical Institute, and a Fellow of the American Statistical Association.
Annette Scherpenzeel is a psychologist and methodologist (PhD) with experience in both fundamental scientific research and in applied survey research. She worked at different universities in the Netherlands and Switzerland, but also in telecom marketing research and data collection institutes. Her special interests lie in the domain of longitudinal research, more specific in sampling, response and attrition, and in innovative data collection methods. She stood at the cradle of the Telepanel in Amsterdam and the Swiss Household Panel in Neuchatel and was responsible for the start and management of the LISS panel in Tilburg, a new online panel for survey research, which was based on a probability sample of the population. The past 10 years, Annette has been a survey manager, responsible for the organisation and data collection of large scientific studies. Among the studies she coordinated were many experiments involving new and advanced methods of data collection, such as Internet weighting scales, GPS tracking, accelerometers, and ecological momentary assessment by smartphones. Since September 2014, Annette puts her expertise in longitudinal household studies in the management and methodology of the Survey of Health, Ageing and Retirement in Europe (SHARE), in Munich, Germany. She is the International Coordinator of this survey since 2017.