Projects / Programmes
Integration of mobile devices into survey research in social sciences: Development of a comprehensive methodological approach
| Code |
Science |
Field |
Subfield |
| 5.03.00 |
Social sciences |
Sociology |
|
| Code |
Science |
Field |
| S274 |
Social sciences |
Research methodology in science |
| Code |
Science |
Field |
| 5.04 |
Social Sciences |
Sociology |
survey methodology, mobile web surveys, data quality, sensor data, mobile technologies
Researchers (16)
Organisations (3)
Abstract
This project will address the integration of mobile devices into survey research, which is one the most promising recent developments in survey methodology. Mobile web surveys, which are designed for completion on smartphones and tablet computers, offer improvements in survey data collection in terms of access to research participants and enhancing survey answers using specific features of mobile devices (e.g. taking photographs, recording audio and scanning barcodes). They also allow data integration from sensors built into mobile devices to provide further insights into an individual’s location, movement, activity, environment and more. However, the exploitation of these potentials in the social sciences remains scattered and limited.
The project will combine a range of methods to conduct research in the following areas:
Evaluation of representation issues of mobile web surveys. We will analyse the characteristics of mobile device users to understand which population groups are suitable for administration of mobile web surveys. We will then evaluate the willingness to participate in mobile web surveys and the effectiveness of various measures used to improve participation, particularly when collecting various types of sensor data from mobile devices.
Elaboration of mobile web questionnaire design principles. We will systematically evaluate problems with questionnaire presentation on mobile devices, where appropriate adjustments are needed to ensure response quality and comparability with web questionnaires for personal computers. Specifically, we will address the methodological performance of enhancing survey answers with photographs and other specific features of mobile devices.
Assessment of integrating sensor data to enhance survey data and improve data quality. We will study the methodological performance of sensor data from mobile devices regarding the user’s location, environment (noise and light) and activity (movement). We will also focus on the potential of sensor data for tailoring survey procedures and understanding the influence of a survey context on data quality. This will also serve as a proof-of-concept for integrating sensor and survey data for other applications in substantive social science research.
The systematic methodological elaboration of mobile web surveys achieved by these project objectives will present a pioneering contribution to the field. It will allow the establishment of comprehensive methodological guidelines and recommendations for the implementation of mobile web surveys in the social sciences. All developed technical solutions will be freely available in the open-source survey application 1KA.
The project results are expected to have a profound impact on survey methodology and social science disciplines that benefit from surveying studied participants in various target environments and enrich survey data with location and other sensor data (e.g. sociology of health, spatial sociology, communication and leisure studies). The outcomes will also be relevant for the development of mobile data collection methodologies in commercial research and official statistics.
The team from the Faculty of Social Sciences is among the pioneers of web survey methodology and is internationally recognized for its scientific publications, international projects, the central resource of information on web surveys (WebSM.org), and development of the open source survey software 1KA. The project group also includes partners from the Artificial Intelligence Laboratory at the Jožef Stefan Institute, who are among the leading researchers in machine learning and data mining, and the Department of Geography at the Faculty of Arts, who work on location data collection and analysis. We will jointly elaborate the project’s activities and exchange findings with other renowned experts from the University of Michigan, Stanford University, Utrecht University and the leading Slovenian market research company Valicon.
Significance for science
Mobile web surveys offer a new way to participate in standard web surveys, which have already become a commonly used and well-respected scientific research method, and open new possibilities for significant extensions of survey data collection. The proposed research is therefore at the forefront of innovation in survey methodology.
This project will be among the first to develop a comprehensive methodology for the general exploitation of mobile web surveys in the social sciences. It will help improve the data quality of current web surveys, where questionnaire design for mobile devices remains one of the major challenges, and simultaneously support further development of innovative survey approaches based on mobile technologies.
The impacts of established methodological guidelines and recommendations are expected to be high. This is especially true for social science disciplines that could benefit from surveying studied participants in various target environments and enriching survey data with information automatically provided by sensors in mobile devices.
The following specific contributions are expected to have particularly important scientific impacts:
Detailed analysis of non-coverage in mobile web surveys will help researchers make informed decisions about target populations that can be successfully studied using mobile web surveys.
The evaluation of measures to improve participation in mobile web surveys will help reduce the non-response bias, which can seriously affect data representativeness. The findings will also be applicable to other survey modes because the non-response problem is currently one of the most serious in survey research in general.
Investigation of questionnaire design approaches and the use of mobile-specific features to provide survey answers will help improve the measurement quality of mobile web survey questionnaires in terms of both answer accuracy and the possibility of obtaining enhanced feedback from respondents. The former is also relevant for standard web surveys, where more and more respondents are using mobile devices instead of personal computers.
Finally, the activities related to sensor data collection will introduce innovative approaches for tailoring survey procedures in location-based research and improve understanding of the impact of survey context on data quality. The latter is relevant for all self-administered survey modes, where the absence of interviewers reduces control over the survey environment. The conducted empirical studies will also offer a proof-of-concept for innovative data collection approaches using mobile devices.
The project’s scope is closely related to other emerging developments in social science methodology, including analysis of large datasets (‘big data’), measurement based on digital footprints and analysis of activities in online social networks.
Significance for the country
This project’s relevance extends beyond academic research. The obtained knowledge and established methodological guidelines will likely be both relevant and directly applicable to various research areas outside of scientific research that commonly use web surveys to collect data and may benefit from the extended possibilities of integrating data collection capabilities of mobile devices into their research. Such areas could include the following:
Market research. Some commercial research organisations and software developers already use data from mobile applications to better understand their customers. They could use the obtained methodological knowledge to integrate such data with high-quality mobile surveys. The relevance of the project’s outcomes in this regard will be further assured by close cooperation with the leading Slovenian market research company Valicon, which develops the integration of surveys and data from specific mobile applications.
Official statistics and other governmental organisations. These entities have also recognised the importance of using various sources of automatically collected data. A recent example is the use of mobile network data to observe daily migrations. The proofs-of-concept and methodological findings of the proposed research will provide such organisations with the knowledge needed to develop new methods for obtaining more in-depth insights into various socially and economically relevant topics. Some examples could include visitor tracking and surveying in tourism research, use of sensor data to track physical activity relevant for health research and measuring environmental conditions in workplaces.
Vendors and users of wearable devices and lifelogging applications. There is a fast-growing market for wearable devices (e.g. smartwatches and activity trackers), other connected devices and mobile applications that allow daily measurement and logging of physical activity, body parameters and lifestyle. Vendors of such devices (like Garmin, Fitbit, Withings and numerous others) increasingly use these data or provide data access for research purposes. The proposed research will be of direct relevance for such research purposes, where combining sensor data with the subjective feedback of users through survey questionnaires can provide significant added value.
Technical solutions developed for research purposes will be freely available as a prototype extension of 1KA software, which is used by thousands of researchers at academic, governmental and commercial organisations (including the Statistical Office of the Republic of Slovenia, the National Health Institute and the Search Results Chamber of Commerce and Industry of Slovenia). Users will therefore be able to directly implement high-quality mobile web surveys and utilise innovative possibilities for extended data collection in practice.
Most important scientific results
Final report
Most important socioeconomically and culturally relevant results
Interim report,
final report
