[Directed acyclic graph building and analyzing based on DAGitty]

Directed acyclic graph (DAG) is a common tool in structural causal models. By visualizing hypotheses about causal relationships between variables via nodes and edges, they help researchers identify bias sources, determine key variables and adjustment sets, and enhance the transparency of causal inference. However, as more variables are incorporated into a DAG, inter-variable relationships often become complex, forming large-scale multi-variable DAGs that pose significant analytical challenges. DAGitty, a browser-based platform integrating DAG drawing and analysis, supports DAG construction, identification of biasing paths, and automatic derivation of minimal sufficient adjustment sets, can greatly improve research efficiency. This paper systematically introduces DAGitty, clarifies its core functions and practical applications to provide reliable tool reference and standardized guidance for research of causal inference.