First: What the frig is selection bias? If participation in research is non-random, then our analytic sample will differ from our target population of interest. This sample-population mismatch can result in selection bias, potentially messing up all our lovely causal inferences.

For instance, if both our exposure and outcome predict participation (ie, collider bias), this can bias the relationship between these variables. (Image from doi.org/10.1038/s414...)