"Respondents not reachable on a home telephone (30% of likely voters) were shown a questionnaire on their smartphone, laptop or other electronic device." . . . . .