Testing for normality in regression models: mistakes abound (but may not matter)

Publication
Royal Society Open Science 12:241904