Methodenzentrum

08.11.2022

New Publication: Causal definitions vs. casual estimation: A reply to Valente, Rijnhart, and Miocevic. Brandt, H. (in press)

In this commentary, I demonstrate how standard mediation analyses may fail to reveal the relevant process variables in single case experimental designs. Using empirical data and simulated data, it can be shown that methods using assumptions about unobserved confounders will produce statistical artifacts in virtually all empirical settings. 

 

Brandt, H. (in press). Causal definitions vs. casual estimation: A reply to Valente, Rijnhart, and Miocevic (2022). Psychological Methods.

Zurück