Minne Hagel

External LIFE Fellow since 2024, Humboldt-Universität zu Berlin

Having attained both my bachelor's and master’s degree in psychology from Freie Universität Berlin, I currently work as a pre-doctoral fellow in Manuel Völkle’s research group for Psychological Research Methods at Humboldt-Universität zu Berlin.
During my B.Sc. and M.Sc. studies, I worked at the Socio-Economic Panel (German Institute for Economic Research), in the Methods and Evaluation / Quality Assurance unit at Freie Universität Berlin headed by Steffi Pohl, and in the Mental Health Monitoring Unit of Robert Koch Institute. Thereby, I got introduced to causal inference in evaluation research as well as the breadth of methodological questions that arise when working with panel data. The question that piqued my interest the most is inherent to many psychological studies but not always explicitly discussed: “How can we ensure that a given statistical effect can be interpreted as causal?”. Michael Eid’s supervision of both my bachelor’s and master’s thesis initiated and fostered my interest in latent variable modeling. In my master’s thesis, I evaluated the statistical performance of the robust chi-square test in Mplus by applying it to different multilevel confirmatory factor analysis models for multitrait-multimethod data.
Now, I am very fortunate to combine questions on causality in non-experimental studies and latent variable modeling in my doctoral project, which focuses on causal effects in latent state-trait (LST) models. Under the guidance of Manuel Völkle and Christian Gische, I will compare different ways of estimating standard errors of causal effects as well as different theoretical perspectives on causal identification in LST models and explore causal effects in continuous time models.

Dissertation project:

Causal inference in latent state-trait models: Standard errors, identification, and a continuous time perspective

Publications

Hagel, M. L., Trutzenberg, F. & Eid, M. (2024). Applying the robust chi-square goodness-of-fit test to multilevel multitrait-multimethod models: A Monte Carlo simulation study on statistical performance. Psychology International, 6, 462–491. https://doi.org/10.3390/psycholint6020029