All time references are in CEST
Wearables, Apps and Sensors for Data Collection 2
|Session Organisers|| Dr Heidi Guyer (RTI International)
Professor Florian Keusch (University of Mannheim School of Social Sciences )
|Time||Thursday 20 July, 14:00 - 15:30|
The recent and ongoing proliferation and development of mobile technology allows researchers to collect objective health and behavioral data at increased intervals, in real time, and may also reduce participant burden. Wearable health devices can measure activity, heart rate, temperature, sleep behaviors and more; apps can be used to track behaviors- such as spending, transportation use or health measures- as well as for ecological momentary assessment; smartphone sensors have been used to capture sound and movement, among others. The COVID-19 pandemic brought about additional uses of apps and sensors to measure population trends on a large range of topics including mobility, access, symptoms, infection, and contagion. Large national studies such as the UK Biobank study and the U.S. based NIH All of Us research program have demonstrated the scalability of integrating wearables in population-based data collection. Other studies, smaller in scope or sample, have developed innovative approaches of integrating apps and sensors in data collection.
However, researchers using these new technologies to collect data face many decisions about which devices to use, how to distribute them, how to process the data, etc. These decisions impact other components of the research design including selection bias and data quality. In this session, we invite presentations demonstrating novel uses of wearables, apps, and sensors for data collection as well as potential barriers or challenges. Presentations may be related to measurement, consent, data storage, data analysis and data collection.
Keywords: data collection, wearables, survey apps, sensors, measurement
Dr Ruben Bach (University of Mannheim) - Presenting Author
Dr Henning Silber (GESIS - Leibniz Institute for the Social Sciences)
Professor Matthias Schonlau (University of Waterloo)
Dr Jette Schröder (GESIS - Leibniz Institute for the Social Sciences)
Mr Frederic Gerdon (University of Mannheim)
Professor Florian Keusch (University of Mannheim)
Technological advancements in the recent past made it possible for researchers to collect and analyze large amounts of health data at unseen scale and speed. For example, fitness trackers and smartwatches produce steady flows of information on individuals' health. Biomarker data and medical records allow to study individuals at levels of unprecedented granularity. Pandemic times when healthcare systems and societies were pushed to their limits have highlighted that access to such data for health research and evidence-based public policy decision-making around the world is essential. However, having access to data depends on individuals' willingness to share their data with others. To better understand the factors that influence individuals to share their biomarker, health, and sensor data, we fielded a survey in the German Internet Panel, a probability-based online panel survey, in early 2022. Using a vignette survey experimental design, we study the impact of data type, recipient, and research purpose on respondents' willingness to share their data. To understand motives underlying respondents' answers, we prompted respondents to explain why they were willing or unwilling to share their data using a set of open-ended survey questions. In this presentation, we will focus on the answers to the open-ended survey questions. Our analysis of the textual answers (n=3,900) is based on natural language processing (NLP) techniques using transformer models. After developing a coding scheme using a subsample of about 500 responses, we fine-tune our NLP model to classify the remaining responses into one of eight categories that capture motives underlying respondents’ answers. Results of this analysis allow us to get a detailed understanding of the concerns that respondents may have when sharing different types of health data with third parties.
Dr Bence Ságvári (Center for Social Sciences, Corvinus University of Budapest) - Presenting Author
Dr Attila Gulyás (Center for Social Sciences)
Dr Bence Kollányi (Center for Social Sciences)
Within the framework of the Artificial Intelligence National Laboratory (MILAB), a pilot research and development project "Using Telecommunication and Sensor Data to Understand and Predict Social Behaviour" was launched at the Centre for Social Sciences (CSS) in Budapest in the last quarter of 2020. The main goal of this project is to develop the 'Octopus Research Tools' (ORT), which aim to create a unified research tool for app-based hybrid (active and passive) data collection via smartphones
ORT is a modular and highly customisable software that can serve a wide range of academic research needs. It allows the combination of different data collection modules, such as longer surveys, time- or location-based short surveys and quick questions, geolocation, and information on device and app usage
We are currently working with several internal academic research projects at CSS that use our tool for their data collection. (e.g. collecting socio-metric network data from volunteers working at the biggest festivals in Hungary; exploring the social class structure in Hungary by combining surveys and digital behavioural data; measuring the working time of freelancers through randomised low-intensity interactions)
The main objective of the presentation is to (1) demonstrate the general principles behind the development of ORT, (2) introduce its basic features, and (3) explain the main challenges related to the technology, convincing potential data providers, and complying with research ethics and privacy standards. We are seeking feedback from researchers working with similar tools and looking for partners for joint research and experiments. More information about the software and the project: www.octopus-research.hu (The website will be updated in mid-January 2023).
Ms Patricia A. Iglesias (Research and Expertise Centre for Survey Methodology, Pompeu Fabra University) - Presenting Author
In recent years, there has been increasing interest in the use of mobile devices to answer web surveys, particularly due to the sensors included in them. This presentation focuses on the camera in smartphones and tablets.
Asking respondents to answer web survey questions by sharing images is expected to present benefits, such as reducing measurement errors for questions where respondents do not know the answer, or improving the satisfaction of the participants, especially when the images can replace several/complicated survey questions. Additionally, images are expected to provide new insights that cannot be obtained through conventional survey questions.
Some research has already been implemented asking respondents to share images. However, very little is known yet about the participants' levels of participation and satisfaction when asking for several photos of objects that are in the respondents' house, in different contexts and countries.
This study helps filling this gap by providing new empirical evidence from a survey asking adults with children in primary school in Spain to take and share photos of all the books that they have in their main residence. These pictures will be used to compute the number of books that participants own, which is a common indicator in the literature to measure cultural capital. Moreover, these pictures will also be used to extract other information, such as what are the main ways in which participants organize their books.
In this presentation, I will report about respondents' participation (e.g., how many respondents send images, how many are in line with the questions) and satisfaction (e.g., did they like it, did they find it difficult) when asked to share such photos. Different types of problems (e.g., technical problems) will also be discussed.
Data will be collected in early 2023 through an online opt-in panel in Spain.
Dr Elina Page (USDA Economic Research Service) - Presenting Author
Ms Lauren Miller (USDA Economic Research Service)
Ms Linda Kantor (USDA Economic Research Service)
Dr Mark Denbaly (USDA Economic Research Service)
The U.S. Department of Agriculture’s National Household Food Acquisition and Purchase Survey (FoodAPS-1) was the first nationally representative survey of U.S. households to collect unique and comprehensive data about household food purchases and acquisitions. This included foods from food-at-home retailers (e.g., supermarkets, grocery stores, and farmers markets), food-away-from-home establishments (e.g., restaurants, fast-food vendors, and schools), and foods obtained for free (e.g., from food pantries, community centers, and from family and friends). Development of the survey's second round, FoodAPS-2, is underway and its design and data collection protocols draw on the lessons learned from FoodAPS-1, advancements in how people acquire food, and both methodological and technological innovations in the surveying environment. That is, FoodAPS-2 will take advantage of web, mobile, and other digital technologies to combat concerns associated with data quality, including nonresponse and underreporting, respondent burden and fatigue, and significant backend data processing times. This presentation will give an overview of the current plans for FoodAPS-2, which were recently evaluated in a large-scale Field Test. We'll highlight the key survey design features, present an in-depth look at a native smartphone application that will serve as the primary mode of data collection, and discuss plans for leveraging extant databases in real-time to reduce burden and improve quality.
Ms Margaret Moakley (RTI International) - Presenting Author
Dr Heidi Guyer (RTI International)
Dr Stephanie Eckman (RTI International)
Ms Leenisha Marks (RTI International)
Mr Carlos Macuada (RTI International)
Mr Jesse Lopez (RTI International)
Background: The All of Us Research Program (AoU) is a research initiative led by the United States’ National Institutes of Health (NIH). AoU collects its data through a nationwide, longitudinal cohort; currently there are over 400,000 participants. AoU participants share electronic health records (EHR), complete surveys, and submit bio samples, all in contribution to the vast AoU database. These data are available to registered researchers to support a wide array of health research. In addition to data accessibility, AoU is also focused on being inclusive of populations traditionally underrepresented in research. AoU participants are diverse in gender identity, sexual orientation, socioeconomic status, education, disability, and health status. In 2019, AoU expanded data collection to include wearable fitness trackers. In their Bring-Your-Own-Device (BYOD) approach, participants who already owned a FitBit device could link their FitBit data to the AoU database. This paper will analyze the demographic representativeness of AoU’s wearable data collection effort.
Methods: Data have been obtained from the AoU database and the analysis will use one-sample chi-square tests to compare the demographic makeup of the participant sample with Fitbit data to the U.S census population and then to the full AoU cohort.
Anticipated Impact: AoU represents an innovative venture into diverse, large-scale cohort studies. The addition of wearables data has the potential to provide key insight into lifestyle factors to compliment what is derived by surveys or EHR alone. However, it is important that the introduction of this data collection component not dilute the program’s commitment to diversity and representativeness in research. This analysis will help to better understand the potential limitations and equity implications of AoU wearables data and the use of wearables in multifaceted, longitudinal studies more generally.