Another backdoor path in bonus example of Lecture 6?

[matt-bledsoe]

Is the path X <- S <- A -> Y not a path that should be considered in the analysis of the DAG in the bonus content of Lecture 6?

[gah-bo]

I also think so...I'll add that 29:15 in the video states a tip to start with: identifying all paths that connect X to treatment Y, with no regard for DAG arrow directionality while doing so...I found that, unless I am missing something, a better "starting point" for me is to identify all paths that connect X to Y, while ensuring I always start at X and end at Y. This is probably what was meant, but it was not obvious to me until I realized that (I think) that path you're mentioning is missing.