HHU - Social Media based Measurements of Psychosocial Phenomena_ Potentials and Pitfalls

Digital Trace Data from online platforms and digital devices, including social media data, has great potential in informing social research about human attitudes, behaviors, and characteristics. Such data is available in high volume and velocity and can complement traditional forms of data in the social sciences, such as survey data, especially in conjunction with computational models designed to work with large-scale data. However, digital traces also have several pitfalls, such as confounding effects of the platforms. In this talk, we critically reflect on some of these pitfalls and devise ways to identify them.

Bio: Indira Sen is a 4th-year Doctoral Researcher at the GESIS Leibniz Institute for Social Sciences. She works at the intersection of Computational Social Science and Natural Language Processing, with a master's and bachelor's in Computer Science. Her research interests include developing mixed-methods to measure attitudes and behaviours like sexism and political approval from new data sources such as web and social media data. Her research has been published in premiere social science and computational science venues such as Public Onion Quarterly and the Conference on Empirical Methods for Natural Language Processing. In the past, she has held internships at Nokia Bell Labs, EPFL, Switzerland, and NTU Singapore.