Originating from genetic studies and psychology, structural equation modeling (SEM) encompasses a broad array of models, which are increasingly being used in applied econometrics as an alternative way to look at causal inference. SEM allows to estimate the effects of a set of variables acting on a specified outcome via multiple causal pathways, and/or to model complex, simultaneous relationships on observed as well as unobserved data. This presentation will introduce basic concepts of SEM modelling – different types of variables and effects, differences between measurement and structural models – as well as practical implementation in Stata and Mplus.