The recent election of Joseph Biden to the White House has raised familiar doubts concerning the efficacy of the polling industry in being able to accurately predict the outcome of an election.
The first response to this complaint could be that it is not the pollster’s job to predict the future. However more specific factors are often raised. One of the most common of these is the existence of the “shy voter” who is uncomfortable in giving his or her real opinion in the polls due to the fear of being judged unfavourably.
The Archive (AMSR) covering the late 1990’s provides much insight on the subject of “shy tories” in the UK. At the base of this concern in Britain was the 1992 general election, where the pollsters incorrectly predicted a Labour win. Did some Conservative voters mislead pollsters? Whatever the answer, a MORI “Letter to our readers” from 1999, accessed from the Archive, feels the need to defend itself from attacks from John Redwood of the Conservative party. It was argued by some in the Eurosceptic wing of the party that MORI should “adjust” their raw polling figures to account for “shy tories”. The 1992 election had been and gone but the idea of the shy voter had stuck. MORI was also accused of “comprehensively misreading” the electorate’s views in the run up to the 1999 European and local elections. Perhaps at play here was the idea that polling not only could help predict the future but also somehow influence it; an idea that brings us nicely across the Atlantic to current times.
With the 2020 US presidential election in mind, it does not seem a bold claim to suggest that perhaps the shyness in “shy voters” (if ever it was there) has been replaced by anger. Poll response rates in the United States have been in decline in recent years; you need to dial over 15,000 phone numbers to get 950 responses. It could be that on top of this, there is a non-response bias from angry Trump voters who see polling as part of the establishment media, ascribing a power to the practice that it should not be worthy of. This implies that perhaps some more focus is needed on non-respondents. In an increasingly polarised world, those that say “no”, and the manner in which they do so could have more and more predictive power than previously imagined.
Sources:
For further fascinating insights into cultural changes and attitudes in Britain, search the Archive from the main search page.
Contributed by Jacob Bartram (BVA-BDRC)
Date posted: 8th December 2020