This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more here

Tuesday, 29 July 2014

Are you thinking what I'm thinking?

Why research needs to go beyond the individual’s perspective to get closer to the reality of behaviour. By Rob Egerton and Jeanette Kaye.

Gestalt rooms, image sorts, debates, projection and elicitation: as market researchers, we’ve become brilliant at exploring the depth of human perceptions and behaviour. We pride ourselves on challenging respondents and finding new ways to understand the workings of their minds and how they make decisions.

However, two important theories – which many of you may have heard of – shed new light on how we can better access the reality of how people behave and challenge some of our orthodox approaches. These theories give us direction on how we can get better estimates of future intentions but also suggest that we should move away from just exploring why the individual respondent did what they did, but rather explore more widely how the dynamics of others around them influence what they do.

The Wisdom of Crowds

The 2004 publication by James Surowiecki raises the interesting notion that group decision-making or estimation is much more accurate than individual decision-making. He uses a great example of individuals guessing the weight of an ox; each individual guess is often likely to be way off but when an average is taken from a group of responses (importantly from what Surowiecki defines as a ‘wise crowd’), the ‘guess’ becomes much more likely to match the actual weight.

“We shouldn’t just focus on how the individual respondent intends to behave in the future but rather how they think others will behave”

This theory has been taken further by ICM’s Martin Boon in his IJMR paper, where he translates this into a new way of conducting election polls – highlighting the superiority of predicting election results via an averaging out of the guesses of a randomly selected sample, rather than the traditional ICM polling methodology. Importantly, this approach had two key distinctions:

  1. Rather than asking each of their polling sample how they intend to vote, the sample is asked how they think others will vote, thereby removing the inherent bias that exists when certain respondents hide their true intentions.
  2. They provided the sample with the previous election’s results. Strictly speaking, this is a deviation from the pure wisdom of crowd’s ideal but it provides respondents with a useful context.

This method was shown to be more accurate than their traditional polling method – and, when you consider that voting intention is notorious for how difficult it is to get a truthful and honest representation, this is quite an achievement.

Surely this same theory also applies to how we explore future thinking and behaviour in the healthcare world. At HRW, we conducted an experiment and trialled two different approaches with the aim of measuring uptake of a new treatment in renal cell carcinoma. Respondents – all of whom were oncologists – were asked how many of their patient cohort would receive the new treatment and, in replicating the wisdom of crowds approach, respondents were also asked how much they thought others would use the treatment.

The first question generated a response of 43%; the second 38%. The results clearly show that respondents revised their estimates when the question was re-phrased. While we don’t have the market share for this hypothetical scenario, we could assume that respondents were overstating their prescribing intentions in the traditional question format, while the wisdom of crowds approach provides a more accurate basis.

As with voting intention, asking someone how much of a product they will use often leads to an inaccurate answer so market researchers often use weighting factors to take account of this. Asking about how they think others will behave removes the need for weighting factors; there is no need for respondents to misrepresent their own behaviour because they’re discussing how others will behave.

So what’s the implication? Both the original work produced by ICM and the study above suggest that whenever exploring future behaviour, we shouldn’t just focus on how the individual respondent intends to behave in the future but rather how they think others will behave. Collating these responses from a wide crowd is likely to therefore be more accurate.

The theory of group behaviour

The second theory, which challenges our traditional focus on exploring how the individual makes decisions, is that of group behaviour. In his book ‘I’ll Have What She’s Having’, author Mark Earls makes the claim that in determining how decisions are made, the influence of other people is often far more significant than the individual’s process of weighing up what they should do.

“This is nothing new”, you might say – but in all honesty, how often do we take this into account when we explore behaviour in the healthcare world? We’re great at understanding the individual’s thought process and what they think, and why they behave in the way that they do but rarely do we map out and understand how customers behave as a group and the extent to which they are influenced by others.

“While it’s difficult to develop a complete picture of how others influence an individual, we should be starting to explore it”

Indeed, recent behaviours to which we can all relate point to how individuals can be encouraged into actions not by their own assessment of what they should do next, but by the actions of those around them. The London riots of 2012 are a clear example; individuals didn’t cause mayhem because they each independently wanted to but because of the actions of everyone around them. Earls also raises the interesting behaviour of laying flowers at traffic accidents or at significant events. It’s a behaviour that we primarily adopt as a result of the actions of others.

We also often explore whether respondents fall into the camp of an early adopter or whether they wait for others to try first, but this is quite a surface-level question and doesn’t really account for the fact that how someone is influenced by others (and whether they have the confidence to try a new behaviour) will be different for each decision that they make.

Group dynamics are complex. The talk given at a recent TED conference by Dereck Sivers provides a refreshing and interesting breakdown of group behaviour, which suggests there is more to it than we might initially think. Interestingly, the video highlights how important it is to break down the behaviour of early adopters versus followers, and there is a greater complexity than we might think as to how we develop behaviours as a group.

So if our healthcare customers are so dependent on others in the decisions that they make, how can do more to understand this through research?

Acknowledging the limitations of just relying on exploring an individual’s behaviour is the first step, but there is more that we can do. While it’s difficult to develop a complete picture of how others influence an individual, we should be starting to explore it. For example, exploring what others have told that respondent, how they have seen others behave and what they have told others, are all valid questions which will help us start to map this journey. It’s also important to look not just at first adopters but also at first followers, thereby breaking down the adoption ladder even further.

So, next time you’re wondering about how best to understand your customers’ decision-making or wanting to know how your customers will behave in the future, ask not what that customer will do but what they think others will do – and focus less on that individual customer but also their ecosystem and those around them.

Rob Egerton is associate director and Jeanette Kaye is deputy MD at HRW

Follow us on
Follow us on Twitter

Readers' comments (1)

  • Does this article apply to understanding users' current processes and behaviours? Everything I've learnt supports that behavioural observation trumps user opinion.

    Unsuitable or offensive? Report this comment

Have your say

Please add your comment. You can include links, but HTML is not permitted.
Your email address will not be displayed on the site. All comments are moderated.

Mandatory
Mandatory
Mandatory
Mandatory