After the Scottish referendum the myth spread that young voters had been betrayed by No-voting older people. The media eagerly seized on the only quasi-exit poll available, one done by Tory peer Lord Ashcroft, and reported that 71 per cent of 16- to 17-year-olds had voted Yes. Organisations like the BBC repeated the “results” of the poll.
Radio 4’s More or Less statistics programme revealed a few days later that the claim was based on an “exit poll” of 2,000 people carried out by telephone after the polls closed. Pollsters asked people which way they had voted – and other questions, including their age.
Only 14 of those asked were aged 16 or 17. Of these, 10 voted Yes and 4 voted No. And yes, 10 out of 14 is 71 per cent. But statistically a sample of 14 cannot be regarded as representative. To publicise this result without a “health warning” attached was sloppy.
More detailed polling before the referendum indicated a clear No majority among the age group.