en
Menü
Zur Übersicht
Overview and Organization

Background, objectives and information on Research Areas

ENTAILab

Research and Innovation Lab: The programme’s core infrastructural service and research center

Projects

List of SPP2431-Projects

The Infrastructure Priority Programme “New Data Spaces for the Social Sciences” (SPP 2431) deals with a topic of paramount importance: It aims at opening up new avenues for the social sciences by building on the strengths of existing panel studies, by innovating new methods and procedures, and by complementing them with the intelligent integration of data derived through new technologies.

It is managed by the programme committee, consisting of Prof. Dr. Cordula Artelt (spokesperson) and Prof. Dr. Corinna Kleinert (both LIfBi), Prof. Dr. Reinhard Pollak (GESIS), Prof. Dr. Stefan Liebig (FU Berlin) and Prof. Dr. Alexander Mehler (Goethe University Frankfurt).

Expanding the analytical potential of existing longitudinal survey data

The programme aims to address emerging fields related to social science research and surveys. This is based on the conviction that the social sciences can only provide empirical evidence to address current global societal challenges if they expand the analytical potential of existing longitudinal survey data through innovative methods and the additional use of novel data sources. Developing new data spaces is an enterprise that requires synergies between different research fields, methodologies, and approaches.

Achieving substantial progress in expanding the scope and strength of panel studies in Germany through various disciplinary perspectives, foci, and approaches

Together, the SPP projects have the goal of researching and creating technical and methodological solutions for the future sustainability of panel surveys, enriching them with data from other sources, and thus paving the way for social science research on key societal challenges. By combining various disciplinary perspectives, foci, and approaches, the SPP aims to make substantial progress in expanding the scope and strength of panel studies in the social sciences in Germany. A distinctive feature of the SPP is its problem-centered approach, which brings together researchers from different fields to work jointly on specific research questions. By using prototypes of existing technologies and developing them further to serve the aims, constraints, and data protection needs of large-scale panel studies, we create synergies. By exploring novel methods and approaches, we establish new avenues of research and build bridges between previously isolated fields of research (e.g., computer science and survey methods). Panel studies will benefit from this exploration of promising methods and technologies that have the potential to reduce the burden of conducting surveys and to systematically enrich data spaces by offering new types and qualities of data.

Addressing theoretical and methodological challenges of integrating different data types, modes of data acquisition, and respondent-driven designs

We see a strong need to anchor the use of data-intensive methods and corresponding analytic methods in related theories from basic research. And there is a need for proof of concepts for integrating existing and generated data types (register and administrative data, Big Data, synthetic data) into social science panel studies and data landscapes. To harness the full potential of innovating social science practices and methods in the long run, the SPP brings together researchers from different fields (e.g., assessment, computational humanities, data science, psychology, survey methodology, and the broader social sciences) to address the theoretical and methodological challenges of integrating different data types, modes of data acquisition, and respondent-driven designs in the field of social science panel studies and data landscapes. Method development plays a prominent role in the SPP because of the importance of new types of data. This includes the integration of methods of natural language processing (NLP) and text mining with methods of multimodal computing, the further development of machine learning methods for recognizing, classifying, and linking instances of these data types, as well as methods aiming at transparent, extensible, and interoperable procedures that produce explainable, comprehensible, and interpretable results.

Answering new and highly relevant research and implementation questions – our four research areas

1) Exploration and integration of different data types

Research on exploration and integration of different data types unleashes the information potential in existing or newly established data collections and amplify this potential by combining and integrating these different data sources. Researchers are thus enabled to better assess both the strengths and weaknesses of different data types as well as their potential to open up new data spaces for substantial research questions in many fields of the social sciences.

2) Respondent-driven designs

Research on respondent-driven designs aims to increase the sampling quality of future survey data in cost-efficient ways, analyzes the potential of tailor-made approaches to and instruments for individual respondents, and establishes new methods of including hard-to-reach populations in existing survey programs.

3) Instrument validity

Research on instrument validity addresses fundamental questions on the quality of measurements. This research detects and overcomes biases inherent in the methods used for analysis, develop new methods to assess and improve the generalizability of its results, and creates new measures and methods for the assessment of existing or emerging phenomena in society.

4) Multimodal data acquisition

Multimodal data acquisition pushes the boundaries of data acquisition. Starting from already established or emerging survey modes, research in this area aims to innovate new modes of data acquisition and to combine various reactive and non-reactive measures in the acquisition of data to open up new data spaces for the social sciences. This includes the development and testing of VR-based applications for data collection based on multimodal interaction (e.g., on eye movement, body movement, and gestures), the application of interactive machine learning (iML), and the use of avatars, for example, as virtual interview partners, which enable automated survey scenarios.

Our 4 Research Areas

For further informations see full proposal of the Infrastructure Priority Programme (→ Downloads).

As a whole, the SPP develops and introduces new data types and make them accessible for research, and it develops corresponding methods for collecting and processing survey data and integrating the various data types.