Beschreibung des Workshops
Dr. Rafael will give a workshop on causal inference. “Correlation does not imply causation.” Whereas virtually everybody agrees with this statement, the more important question lies in how you establish a causal relationship. This is crucial for testing and advancing theory because most theories are built on causal relationships, and science can only progress when theories are robustly tested. Yet, many published articles in top tier journals fail to address potential threats to causal inference (e.g., Antonakis et al., 2010). In this workshop, we will discuss 1) what is causality, 2) different threats to causality, such as omitted variable bias or omitted selection, and 3) estimators and research designs that can help establish causal relationships. We will use different R packages that will be communicated before the workshop. The following topics will be covered:
- Introduction to causal inference
- Quick review of Ordinary Least Squares regression
- The endogeneity threat and its forms
- Alternative estimators to establish causal relationships (e.g., Instrument variable regression, regression discontinuity design, or difference-in-difference regression)
Zielgruppe
Promovierende des PK NRW, die sich für ihr Forschungsvorhaben mit Causal Inferences auseinandersetzen und bereits Vorkenntnisse in Regressionen (OLS) haben.
Dauer des Workshops
1. Tag: 09:00 – 17:00 Uhr
2. Tag: 09:00 – 17:00 Uhr
Anmeldungen an uum@org.pknrw.de bis zum 25.03.24; Mindestteilnehmeranzahl 6
