ABSTRACT
R.A. Fisher (1938) once wrote: “To consult a statistician after an experiment is finished is often merely to ask him to conduct a post mortem examination. He can perhaps say what the experiment died of.” Fisher’s words apply equally well to surveys: implementing a badly designed survey can be worse than collecting no data at all. Conclusions drawn from a poorly designed survey, such as a call-in poll in which individuals volunteer to be in the survey, can be completely misleading: all that a survey statistician can do after the deed is point out the design flaws that make the results questionable or false. One example where lack of attention to survey design relative to intended use led to possibly erroneous conclusions occurred in 2002 when CBS News told Americans that “Sleeping longer-like getting eight hours or more a nightcould shorten your life” (https://www.cbsnews.com/stories/2002/02/14/- health/main329440.shtml).
