Tuesday, April 5, 2011

On changing human behaviour

Changing human behaviour, so as to get people to act in a manner that increases the likelihood of achieving certain social or environmental goals, is one of the most challenging areas of public policy. Standard approaches involving regulation and incentives (taxes, rewards, and penalties), while effective to certain extent, are increasingly becoming blunt instruments, especially on the more universal of social and civic problems.

In this context, Richard Thaler and Cass Sunstein's path-breaking book, Nudge offers several interesting insights into how human beings can be subtly nudged into performing specific tasks. And a series of small nudges could go a long way towards meeting important public policy goals.

An excellent summary of these techniques comes from Oliver Payne with his 19 ways to "ask" for sustainable change in human behaviour. He has three presentations (Summary, Part I, Part II, and Part III, this, this) form an excellent resource.

1. Simply Ask - eg. when asked nothing in a food Que, only 40% of students took a serving of fruit, but when specifically asked whether they will have fruit, nearly 70% took fruit; voters who were asked a few days before voting whether they will vote were more likely to turn up and vote etc. The "exposure effect" increases the likelihood of the desired outcome.

2. Ask using the right words - eg. carbon offsets are more acceptable than carbon tax; user charges are easier to push through than taxes (framing of the issue); nudge to prevent people from stealing wood from Arizona's Petrified Forest National Park ("Many past visitors have removed petrified wood from the Park changing the natural state of the Park" Vs "Please don't remove the petrified wood from the Park in order to preserve...", the latter was more effective) (reinforcement of social norms); describe carrots as "X-ray vision carrots" (to pre-schoolers) or soup as "Rich Vegetable medley Soup" increases uptake considerably (selective perception) etc.

3. Ask using the right images - eg. the dual image of a littered environment being changed to a clean one reinforces social norms and is more effective in driving home the message on littering than just a littered environment image (it ends up reinforcing the damaging message that many people do litter).

4. Ask using the right authority - eg. Don't Mess With Texas campaign (which did not work with fines) reduced roadside littering by over 70% over 5 years through a campaign with sporting and country-music heroes imploring people to not litter. Ads avoided the negatives of shame and guilt in favor of the positives of pride and group identity (reinforcement of social norms).

5. Ask using the right fake authority - eg. An office tea and coffee "honour box" (into which people dropped the charges) was more effective at boosting honesty and collecting money when a pair of eyes (Big Brother Eyes) was displayed beside it (authority effect - sensitivity to our actions being observed by others); smiley and frowny faces about your driving speed on electronic signboards in South Lanarkshire Council roads (instead of numerical speed information) was more effective at reducing speeds (social approval - smile, you're on camera!); smiley and frowny faces to represent electricity usage on consumers' electricity bills by South California Edison electricity utility's OPower Home Energy Reporting System etc.

6. Ask in the right order - eg. listing disadvantages followed by advantages of carbon tax was found to be more effective in getting public approval than the other way round (framing and anchoring effects)

7. Ask at the right time - eg. traffic light synchronization program in Texas which informed drivers (through digital signboards) about their optimal traffic speeds lowered delays by 25% (self-serving bias)

8. Ask with the right incentive - eg. RecycleBank has a program in many US and UK cities that weighs the amount you re-cycle and converts it into points which can either be redeemed for shopping coupons at local stores or brand outlets (partnership with eBay and Marks & Spencers) or informs them the equivalent numbers of trees saved and oil barrels conserved.

9. Add options - eg. using decoys to help people make choices between various options (by say, adding an additional qualification to the item we want people to purchase or not purchase); keep a non-recyclables (or trash) bin beside the recyclables so as to ensure more effective sorting (framing effect); also small hole for recyclables bin and a larger hole for non-recyclables bin etc

10. Take away options - eg. default options in computer programs nudges data entry operators away from making mistakes; mandatory fields and server clock times too reduces the probability of errors in data capturing.

11. Ask, but have a default option - eg. have a default menu option in school restaurant or conferences which is vegetarian (or healthy food) and provide non-vegetarian (or junk food) when asked (dramatically increases uptake) (framing effect); California's Ready Return tax filling form is pre-filled with last year's data was widely welcomed by assessees etc.

12. Ask a completely different question - eg. Piano staircase and calorie counters on steps (Goodnight Hostel in Lisbon) encouraged people to use stairs over the escalators (framing effect); Bottle Bank Arcade bins placed at strategic locations in Swedish cities that asks people to deposit used bottles and cans.

13. Let the feedback ask the question - eg. Ambient Orb device nudges people to optimize their electricity usage. Cognitively salient information helps people overcome inertia.

14. Don't Ask, Tell - eg. Inform tax payers that evasion is the exception and most people pay or put cards in the toilet to inform guests that most other guests re-use their towels. In both these cases, there is a reinforcement of a social norm.

15. Ask nothing, other than simply to measure - eg. Drivers become more mileage conscious with merely owning a car (say, a hybrid car) whose USP is mileage.

16. Don't ask anything - other than they go public - eg. grading restaurants in Los Angeles (reinforce social norms); Wattson household energy monitor whose data is displayed on the owner's Facebook page (Social norms - peer pressure) etc

17. Ask for a commitment - in the future - eg. shower timer to control water flow (temporal discounting or time inconsistent preferences); commitment contracts on exercising and eating habits on StickK.com etc

18. Ask Kinetically - eg. automatic light and AC on and off when key is inserted or taken off the slot in hotel rooms; square peg, compared to round peg, to hold lavatory paper (each tug is met with a resistance, which encourages people to optimize on their toilet paper usage) etc.

19. Make the question irrelevant - eg. smaller plates to reduce over-eating (selective perception); moving the clock backwards and forwards to make more optimal use of sunlight.

As can be seen from all these, loss-aversion, framing, and social norms are the commonest cognitive biases that can be targeted to formulate policies. People draw different conclusions and act differently based on how the information/data is presented - people are context dependent. People are much more averse to losses than to similarly sized gains - pain of loss is twice the pleasure of gain! People prefer to follow the herd and their actions to reinforce the social norms.

No comments: