New Dimensions in Social Science: Extending Structural Equation Models to Accommodate Novel Data Sources
Using traditional data such as questionnaires, social and behavioural scientists have become highly familiar with Structural Equation Modelling (SEM) as a method to understand relationships between unobserved constructs. However, new technologies generating fundamentally different types of data are rapidly becoming available. Currently, SEM cannot be used with these often complex and high-dimensional data. This project aims to enable their use by (a) incorporating machine learning techniques into the SEM framework, and (b) extending those techniques to leverage substantive theory encoded in SEM. This opens new integrated ways of researching interdisciplinary social science questions.
Dr Daniel Oberski & Prof. Irene Klugkist
NWO Research Talent Grant
1 September 2017 – 1 September 2022