While doing their two weekly national surveys, Rasmussen and his team noticed an anomaly. Out of every 1,000 or so respondents, there would always be three or four who were far more radical than everyone else. After several months of finding these unusual responses, Rasmussen realized they all shared three characteristics.