Polling services were heavily criticised in the aftermath of the General Election in May, 2015 — no one, apart from tech group Survey Monkey, had accurately predicted that the Conservative party would win. Basically everyone underestimated how much support David Cameron’s Conservative party had.
This week, an inquest into why no polling services managed to give an accurate prediction on the General Election is expected to be published.
But couple that with the lead up to the referendum on June 23 — on whether Britain should leave or stay within the European Union — and polls are coming under scrutiny again.
In fact, considering how polls fared in the General Election and how there are notable discrepancies in the “remain” or “leave” camp when it comes to certain polls, it suggests that results should be taken with a pinch of salt.
On Wednesday, the Financial Times pointed out a few key variants to the way polls are conducted, which is why poll results can’t accurately be taken at face value:
Phone polling gives the “remain” camp a lead in the polls compared to online polls — “The Remain campaign that are 15 to 20 points higher than those carried out online.”
How the question is phrased makes a massive difference to responses — the FT spoke to James Kanagasooriam, from polling company Populus, and Matt Singh, a polling analyst who also said “the failure to take into account the wider social attitudes of the people surveyed” can also make a difference to the outcome. The FT pointed out that Singh “looked at 80 EU referendum polls that have been published since last year’s General Election.”
Phone polls don’t have a “don’t know” option — Singh told the FT that phone polls don’t do what online polls do and present an explicit “don’t know” option.
Politicians and the media want quick polling turnaround times — Singh and Kanagasooriam told the FT that since those who commission polls want an extremely quick turnaround, this can mean “they struggle to connect with hard-to-reach sections of society, such as those who are more socially liberal.”
So in other words, not all polls are conducted in the same way and therefore will not be able to carry out accurate readings. However, with the inquest being released soon and the EU referendum in June, it looks like 2016 will be the year of reckoning for pollsters.
In fact, Singh and Kanagasooriam said to the FT that “traditional [polling] modelling may no longer be sufficient to draw an accurate and representative sample of the voting public.”