Shortly before the 1932 presidential election, a national magazine conducted a telephone survey of voters. Based on the results of the survey, the magazine predicted that Herbert Hoover (Republican) would beat Franklin Roosevelt (Democrat) by a landslide. In fact, exactly the opposite happened. The trouble with the survey was that in those Depression days, only rich people could afford phones, and they were likely to vote Republican. This is a classic example of an error in which phase of inferential statistics?

Question