In a month dominated by headlines on spy poisonings and shady firms allegedly scraping Facebook data to swing elections, you could be forgiven for not noticing the announcement of a new firm on the UK opinion polling scene. Deltapoll boasts pedigree, a collaboration between former ICM and YouGov directors. And while its launch may not have usurped the stories above, early signs of Deltapoll’s approach suggest it could explain the unpredictable currents in public opinion that generated them – and where the stories might go next.
You may also wonder why a new pollster is choosing to break the market now. It’s been a torrid few years for the polling industry, hit hard by two general elections in a row where almost every firm called it wrong. In looking for reasons why, factors such as the increasing difficulty of getting through to an accurate cross-section of the population, complicated by the growth of shift work, mobile phone usage, and the famed Home Counties “We’re on the Telephone Preference Service – don’t you dare call me again” response come up.
Beyond those, firms still have to deal with the ultimate challenge of polling: figuring out whether the people they speak to genuinely mean what they say.
The media spent the 2015 General Election campaign bewildered that Labour were still tying with the Conservatives in the polls. The result revealed that the polling error wasn’t far off the proportion of voters who consistently said they would vote Labour but also that they preferred David Cameron as Prime Minister.
In 2017, the polling challenges were twofold. Firstly, determining if the Corbyn surge was real. Secondly, the overnight switch from gauging if Theresa May was competitive in safe Labour seats to gauging if she was still competitive full stop after her so-called ‘dementia tax’ pledge.
In both campaigns, confusion stemmed from a feeling that the topline figures weren’t fully reflecting the gut emotional response to how the public was reacting to the election campaigns.
This is why Deltapoll’s approach is promising. They plan to incorporate implicit response testing into their research, measuring respondents’ unconscious kneejerk responses to politicians, parties and messaging. This will make them the first major UK polling company to use this long-established academic technique as standard in fieldwork. It has rich potential.
Readers of Jonathan Haidt’s The Righteous Mind will be familiar with the metaphor of the rider and the elephant, explaining the roles reason and emotion play in voting behaviour. Decision-making is largely driven by the emotion and instincts (the elephant), with reason (the rider) mostly clinging on for life and providing post-justification for what the elephant already decided. Or, trying to hem away from the elephant’s worst instincts and prevent a massacre of cognitive dissonance.
Using implicit response testing in 2015 might have revealed that many voters who answered Labour on autopilot probably weren’t going to put their cross next to the Labour leader’s name. Or, in 2017 that the early rapturous support for Theresa May in traditional Labour seats was only skin-deep.
Techniques that provoke and measure gut reactions don’t only have a place in polling. They’re a key part of our research at Linstock as well, from commuting habits, to investment behaviours, to how to nudge more people to save for a pension. Often the best way to understand behaviour is to draw out emotional reactions – the moments of frustration, happiness or excitement that often drive action.
Using tactile fuzzy felts to get respondents to think about their commute is one unusual tool we use to explore behaviour. The aim is to provoke deeper thought and get past autopilot answers about driving to work. Techniques like this are often more likely to reveal why people behave in the way they do, highlighting the key barriers to adopting new, preferred behaviours including cycling or taking the train to work.
Deltapoll’s approach looks like it will bring a welcome shot of behavioural science insight to opinion polling. We’ve got a gut feeling the results will be interesting.