Anyone who has spent even a little time asking people questions about money—how they spend it, how they save, how they plan ahead—already knows the responses do not always accurately reflect what’s happening in real life. For a lot of reasons.
MicroSave has given a fair amount of thought as to how we might manage this exchange to achieve less biased results, particularly since much of our research is qualitative. (We are after the reasons, why and how, specific aspects of financial inclusion succeed or fail.) Participatory research methods focus on community rather than individual needs help make respondents less self-conscious and more willing to engage.
Nevertheless, anything involving cash flow and payments, behaviour patterns, or a specific new offering, still elicits a biased response.
The most obvious reasons being that most of us, rich or poor, would prefer to present our money management skills in the best possible light, rather than admit to a researcher—or anyone else—uncertainty, fear, or ignorance. Research participants also tend to over-think the situation and respond to what they believe the moderator wants to hear. We already know, for example, what happens when we ask respondents—no matter how subtly—a hypothetical question about a new cattle insurance offering or better debt management.
Almost all of us went to school. Everyone, including students with only brief and limited exposure to education, learn early that if you tell the teacher what s/he hopes you will say, you are far more likely to succeed in that class, and every other class, than the kids who ask too many questions and promote alternate opinions. (A lot has been written on this topic, but here is a useful summary of what goes wrong in classrooms—and its longer-term effects.)
So, how do we move away from respondents seeking to provide what they believe are the “right” answers? One solution we came up with that seems to be working surprising well is to involve them in games such as Chutes and Ladders whereby decisions, in this case financial ones, have gratifying, or dire, consequences.
But since it’s just a simple game, players can take more risks, feel less chagrined when they fail, and discuss the consequences of their actions more candidly. Our research questions and follow-up probes become part of the game, rather than a discussion about abstractions and hypotheticals. The understanding is better when one is in action [rather] than listening or reading.
Hypothetical situations are also easier to introduce in a simple game, and for respondents to answer spontaneously without the worry of making “mistakes”. In some instances, we even award small prizes like chocolates and pens to game winners, again to shift the focus away from correct or incorrect answers.
Instead of introducing a new form of livestock protection as a new service and asking who might sign up for it, we incorporate the question into one of many decision points in game. As a result, answers now range from “Yes” (generally those who have many cattle and already understand the cost benefits of insurance) to “No” (those who have very few and see insurance as too expensive) to “Not sure” (the many who are unclear about what insurance entails and, in a game, welcome an explanation). We discuss all three answers and then move on with a roll of the dice.
“Games aren’t tests; they’re just fun,” notes Premasis Mukherjee, a senior MicroSaveresearcher. “Everyone gets involved—and everyone is also reminded, very usefully, the important role luck plays in all our lives.”
MicroSave has already used this technique with some gratifying results in our recent studies on information sources and financial capability and financial metaphors. We plan to incorporate it in current and future research efforts as well.
We would be delighted to hear your own thoughts on using games and other informal, appealing techniques that encourage more personal involvement. Our goal is to free respondents from the performance anxieties many experience—and the less than trustworthy responses that result—in too many research situations. These are nevertheless still experiments in progress and others’ findings, positive and negative, are very welcome.