We all know that people sometimes say one thing and do another.
This can be a problem if you are in the business of undertaking surveys.
In the run up to the US presidential election in 2016 the Trump vote was seriously undercounted by pollsters.
One explanation is that people did not want to publicly reveal that they planned to vote for an unstable megalomaniac.
This so-called ‘shy voter’ effect means opinion polls can be unreliable guides to the future.
Something similar is at play when market research touches on subjects that enable respondents to burnish their self image.
In a survey someone might state that they only eat organic food and listen to alternative music, when in reality they enjoy McDonalds and Adele.
This might account for why the Living Costs and Food Survey by the ONS records fewer than half of cigarette purchases made each year.
It might also explain why so many of us express concerns about climate change and the environment while driving diesel cars and filling bins with plastic waste.
Understanding the difference between stated preferences and revealed preferences is a critical lesson in business, as the founder of any successful start-up will tell you.
Asking prospective customers if they will pay for your product is almost a pointless exercise.
“You can’t ask people who haven’t paid how much they’re willing to pay. Their answers don’t matter because there’s no cost to saying ‘yes’ ‘$20’ ‘no’ ‘$100’. They all cost the same – nothing,” explains Jason Fried, CEO of Basecamp.
“The only answers that matter are dollars spent. People answer when they pay for something. That’s the only answer that really matters.
“So put a price on it and put it up for sale. If people buy that’s a yes. Change the price. If people buy, that’s a yes. If people stop buying, that’s a no. Crude? Maybe. But it’s real.
“You can dig into the why’s more deeply over time, but you have to start somewhere. And the best place to start is with real answers.”
Talk, after all, is cheap.