[MRMW] Understanding human behaviour to improve mobile research design

Shirley Eadie, founding member and CEO of Pondering Panda, explored some of the idiosyncrasies of the human condition at last week's #MRMW event, in order to help attendees understand the impact of those idiosyncrasies on research design, to lead to quick wins for better, more accurate mobile research.

Eadie's presentation was very well received as she drew on the vast and varied learnings from the field of behavioural sciences to explore the inaccuracy of memory as well as how an inflated sense of self, irrationality, and our emotional nature wreak havoc when conducting research, as well as how to practically apply what we've learned about the human condition when researching mobile phones.

Starting with a story, Eadie spoke of a group of researchers who went underground to explore whether mobile could solve typical market research problems. She said they conducted thousands of interviews over six months, which was exceedingly complex. Eadie says all Africans need to get over the self-doubt of whether our findings are good enough to share with developed markets, as we're definitely on top of our game and South Africa is on par if not leading the industry.

@shirlswakefield Cracking the code on behavioural science insights for mobile... #mrmw pic.twitter.com/vlDjAzOPUV

- Pondering Panda (@pandainsights) November 5, 2014

Next, Eadie spoke of system 1 and system 2 thinking, as identified by Nobel Memorial Prize in Economics winner Daniel Kahneman's best-selling 2011 book, Thinking, Fast and Slow. Eadie says the book highlights the conscious and unconscious ways we think and process the world around us, such as ways we interpret and constantly fill in spaces - we don't need to see the full picture to do so.

Eadie explains that we think very well linearly, with 84,000 generations of hunter-gatherer behind us, whereas exponential thinking has only been around for 8 generations, so our brains are not yet primed for it. This means that the smarter you are, the more you've learned to think in terms of short cuts and heuristics. We make fast and frugal decisions and aren't good at deliberate decisions, and we think far less than we think we think.

Don't forget that there are always some cognitive biases at play, some hidden forces behind how we think and consume information. There are a number of context, time and social biases that point to errors in terms of research design and how it's completed.

Contextual bias: Choosing the answer because of how the answers are presented

First is contextual bias, based on how something is presented to you, such as the actual words, sounds or framing that alter how it's perceived, like optical illusions. Eadie says this is used commercially, for example, Rolls Royce sell the most cars at boat shows, because "if you're walking around to drop a few million on a yacht, what's a few million on a Rolls?" In another example, Eadie says your perception of a stranger is altered by the temperature of the drink you're holding in your hand so if you're holding a hot drink you'll perceive the person as warmer.

Anchoring is another form of framing with the impact of numbers and how we see other numbers. For example, Eadie says we'll perceive something marked R70 as being less attractive than something that was marked down from R100 to R70 as people like a sense of value, so when it comes to pricing and pricing research, we need to show the options in isolation, not comparatively. Salience is the last contextual bias Eadie mentioned, which is when certain information that stands out and seems relevant affects how we make decisions, such as when order bias, especially in a small feature phone screen, wreaks havoc. Eadie says this is why rotations are vital. Mobile market researchers also need to eliminate the need for scrolling, reduce the character set and response fields, capitalise on large samples and only show one concept per cell.

Time bias: why research should be all about the here and now

Eadie says we're particularly bad at looking back and the further we are from an event the worse we are at remembering how we behaved. We're just as bad at forecasting to the future though, which is why we give more weight to present events than we do to future ones. To highlight this, Eadie showed the well-known Absa Savings 'Marshmallow test', saying we're no better at postponing a reward as adults.

This is why banks and financial institutions try get us to imagine what we'll be like when we're older, in order to connect us to the future selves we struggle to imagine. This highlights the importance of analysing data on time, such as the impact of the time of day a question is asked, as the question "how likely are you to buy a snack" shoots through the roof at lunchtime. Eadie says it's so simple, why aren't we looking at data from a time perspective more often?

Next up was diversification bias. This is quirky, proof of our tendency to think we'll be more interesting in future than we are now, and that we'll make better and more varied decisions, such as trying something new at your favourite restaurant instead of sticking to your favourite meal. Eadie says this ties back to shortcuts and the brain conserving energy. Eadie explained what this means for evolving research design by stating it's essential to date and time stamp everything and keep in mind that mood changes throughout the day. Market researchers should therefore use predictive analytics to tunnel into data due to the sample sizes they can access, in order to simulate future scenarios due to improved computing power.

Social norms: Fitting in to stand out from the crowd...

Eadie touched on social norms as the final cognitive bias to be aware of. This is a type of framing based on the extent to which people in the environment affect us, such as how we slow down to look if see a crowd forming. Eadie says this phenomenon is due to us having so much practice wearing the hunter-gatherer hat. We're used to copying behaviour and asking questions later. Take the stock exchange, fashion and dieting as examples - we believe something to be more correct the more others do it. Interestingly, Eadie points out this means those who try their hardest to be different are also part of a tribe - visit Exactitudes.com, a social experiment in this regard that shows we're all desperately trying to be different but still emulating the environment around us. This is why Eadie says there's such value in engaging with people on mobile, as there's a huge amount of interesting work out there that can be accessed through the privacy of mobile phones.

But there's more to it than that. Don't forget about 'me error', which is the inability to judge your own behaviour as we are much better at judging others. This means instead of asking customers about their decision-making, it may be better to ask the broader audience why they think your customer is engaged with your brand over asking their own opinion. This form of 'me to we' research also has the benefit of allowing for smaller sample sets.

Eadie concluded by stating mobile is ideal for researching socially sensitive topics as it reduces interview bias and group think, as they actually register differently in our brains. While we don't have all the answers, mobile is a fantastically experimental landscape we can use to understand how to better design the research. So while there's "lots of cool technology out there," we need to simplify the process by understanding the human condition. Also keep in mind that mobile itself creates framing, but if we know that we can use it, so market researchers ultimately need to create awareness to reduce these errors more than in the past.

Spread the word and follow @Pondering Panda on Twitter.