It’s hard not to feel that market research enters 2016 with a whimper rather than a roar.
Even before the mixed performance of last year’s general election polling, developments in neuroscience and behavioural economics were prompting some to claim that surveys are bunkum because it’s nigh on impossible for participants to answer questions in a way that accurately accounts for their true motivations and desires. What’s more, research is increasingly alienating participants, as evidenced by declining response rates.
In short, while the traditional survey certainly isn’t dead, it is under pressure. Nowhere is this more evident than in the area of online research panels. These are comprised of people who complete surveys in return for money, and they are a mainstay of general public research.
Many research studies sample only a tiny proportion of their target audience using online panels, be it, for example, the UK electorate, recent car purchasers or people who shop at Sainsbury’s. That’s fine if the sample is skillfully constructed to be representative. Unfortunately, this isn’t always the case as online panels are largely self-selecting (i.e. we have relatively little control over who joins them and who completes surveys, not to mention how many times they participate under different aliases when financial incentives are dangled in front of them.) It’s often not valid to extrapolate the views of a small number of people in these circumstances.
A related issue is the likelihood that more opinionated or time-rich people tend to sign up to online panels in the first place; how then do their opinions differ from those of the majority who are not inclined to become members?
Compounding this challenge, the proportion of people disinclined to participate is increasing. If fewer people are participating in online panels, does that mean we are increasingly left with a hardcore of semi-professional survey takers who are perhaps more motivated by incentive money or simply like taking surveys more than most? If so, it’s hard to make a case for their views being truly representative.
In light of this, we have high hopes for the Vivi online community, which is composed of people who participate in research in order to raise funds for causes they care about. Can this model generate more representative data and more considered responses than those gleaned from a traditionally incentivised online panel?
As yet, we don’t know. These are questions we’ll attempt to answer in 2016. But we believe this is an important moment for research because, even as the limitations of online panels become more apparent, opinion polling seems to be gaining ever greater prominence in the media. This is not necessarily a sustainable situation.
Finally, despite the concerns outlined above, we’d stress that online panels are far from superfluous; they just need to be used appropriately. By painting a canvas of consumer opinion at relatively low cost they continue to play a valuable role. Their brush strokes may be broad and the resulting picture impressionistic, not detailed and precise, but this is often sufficient to meet client objectives.