The divergence between phone and internet polls: which should we believe?

One of the striking features of polling of referendum vote intentions to date has been that polls conducted by phone have been producing more favourable results for Remain than those undertaken via the internet. Inevitably, this has raised the question of which set of polling numbers (if either) should be believed. This analysis paper assesses the evidence produced by the attempts to answer this question and considers what guidance it gives us as to which set of results is more likely to represent accurately the current distribution of voting intentions amongst the general population.

Two ways of polling, two sets of results

  • Phone polls have been producing more favourable results for Remain than those undertaken via the internet.
  • On average, internet polls are suggesting a tight race while phone polls are putting Remain slightly ahead.

Possible explanations

The ‘Don’t Know’ issue 

  • On average, phone polls tend to report a lower proportion of ‘Don’t Knows’.

Shy Leavers?

  • Are some respondents reluctant to say that they are inclined to vote Leave?

Sampling differences 

  • Are phone or internet polls finding too many people in one group or too few in another?

The role of education

  • Should pollsters be reporting information about respondents’ educational background?

 

Read the full analysis paper:

The Divergence Between Phone and Internet Polls: Which Should We Believe?