"can be used to program entirely a candidate's bid for political office. This is no longer some science fiction dream or a
possibility for the twenty-first century. It has already happened."
"For example, if a random sample of 600 people in California
were asked whether they proposed to vote for Reagan or Unruh
in the [California gubernatorial] election [of 1970], the response
would be accurate to within 3 per cent, 95 times out of every 100
samples taken. In other words, if operators telephoned 600 people
at random in California, and went through the same procedure
another 99 times, each time with 600 different people, on average 5
samples could be thrown away, because they would be the bad apples.
The other 95 would be accurate to within plus or minus 3 per cent.
If the sample said that 56 per cent of people would vote for Reagan,
then there was a high degree of probability that the Governor [Reagan]
would receive between 53 and 59 per cent of the vote in the state."
"...It was not simply the fact that he [Reagan] had a long career of acting behind him which allowed him to deliver the right line, mood or expression when required. He had also been a governor for eight years, a candidate four times (five if his 1968 bid for the Presidency was counted) in state and national elections over fifteen years. Regan's six years as a front man with General Electric, in which he had visited 150 plants in 38 states, had made him responsive to and experienced with live audiences so that he was equally at home in front of a camera or a crowd. A combination of acquired skills as a film and TV actor, politician, PR man, candidate, journalist and radio announcer over forty years had made him the most nearly perfect person for programming in the history of politics."