Changing Behaviour - Is it Child's Play?

Many of us who are working for new metrics of progress are interested in changing behaviour for two broad reasons:

1) We believe that what we measure affects what we do - and so we believe new measures will lead to new (improved) behaviour; and

2) New measures will not simply be adopted on their own: we need also to change behaviour if we want society to start paying more attention to the measures we build.

Small children are - quite possibly - the world's most effective (and definitely the most persistent) changers of behaviour. At least mine are when it comes to me and my behaviour. Millions of years of evolution has instilled ruthless strategies within their DNA to help them get what they want. And this weekend I had the dubious pleasure of being subjected to every weapon my two kids could deploy in operation Xbox. Perhaps there are some parallel approaches we in the indicator business can learn.

I can think of at least 5 strategies employed.

1) Please can I have one

Neither convincing nor effective. At best it will provoke a muttered 'We'll see' (or 'we will run a consultation and scoping study'). More likely outright refusal. (c.f. nice idea but we have no resources).

2) Everybody else has one

Again, unlikely to be successful. It may give momentary pause for thought ... but this is almost always immediately followed by reasons why the circumstances are actually different at home. "Well you already have a nintendo, a wii ..." (c.f. we already have a mature set of national accounts). In fact, as few countries do have a set of progress measures the argument is often used by those opposed to the idea.

3) If I don't get one I will make life difficult/ If I do get one I will cooperate

Does anyone ever give into this? Promises of better behaviour might carry more influence but are unlikely to be all that impressive ... and shouldn't they be behaving well anyway? (c.f. if you really want to help improve our measurement system then why not join our statistical advisory council).

4) I need one because it will help me do A, B and C

One of the more convincing arguments, this one, especially if A, B and C are worthwhile in themselves or can help overcome some other problem (its educational.. it will help me have more friends). But while it will often give pause for thought it may not be decisive to push the wavering decision-maker to action. (I'd say most of the arguments included in the work of the Stiglitz-Sen-Fitoussi commission for example fall into this category).

5) If I do X, Y and Z then can we have one?

The weapon of last resort of any scheming child but - depending on the merits of X, Y and Z - is quite often the winning pitch. The benefits that will foreshadow the purchase (or the construction of the indicator set) are sufficient in themselves to justify the investment, even if the game (or new indicators) will not be used as much as we'd hoped.

This last argument is one that I have long been interested in when it comes to indicators and it is one that has been largely neglected by those advocating for change. A variety of partial evidence around the world suggests that the process of discussing and constructing an indicator set can bring important benefits, notwithstanding how - or even whether - the indicators are used. For example, discussing indicators of progress can:

a) Engage citizens in important discussions about their future and help get citizens and NGOs talking more – and more productively - with government (and vice versa);

b) Help different sectors (or even bits of government) agree on the outcomes they are jointly trying to achieve;

c) Help statisticians engage better with users (and vice versa); and

d) Reframe contentious issues in a way that can help find common ground. This in turn can lead to more trust and more constructive debate. (The political right and left may never agree on how to tackle poverty, but they might agree on why poverty is an important issue to tackle and how it should be defined for example, which is a significant step in itself).

In April last year I was at a meeting with Joe Stiglitz in New York and he too noted his interest in the benefits that can flow from the process of indicator construction. Nine months later I'm pleased to say the Bertelsmann Foundation have launched an important research project to look into this more closely. We are trying to develop an armoury of arguments around the benefits of the construction process that will help persuade those sceptical policymakers to take new measures of progress more seriously.

If you know of indicator projects where the process was particularly beneficial, or can think of other ways in which the process can help, please do get in touch because the research is just beginning. And if the arguments that come out are half as effective as those used by my kids we will be one step closer to a more informed public debate. And, yes, the Xbox arrives tomorrow.

Jon Hall