Thursday, March 21, 2013

Nudge Database

Update: Here is the latest version of the Nudge Database pdf

This is a list of empirical ‘nudges’ and their sources with a particular emphasis on those sourced from academic papers. It will be updated regularly. I created it because while there are many interesting papers and websites documenting nudges, there is a lamentable absence of any searchable, central database. If you know of some nudges that I have not included, please let me know by email or TwitterThe nudges listed here are intended as a quick reference: if you are looking for sample sizes and p-values you need to check the original paper. Where possible I have provided links to freely accessible versions of the papers.
Part I || Part II || Part III || Part IV || Part V || Part VI || Part VII || Part VIII || Part IX ||  Part X

1. 
Nudge: Using defaults in organ donation to increase compliance rates. Those countries where people are required to opt-out of organ donation report significantly higher consent than those with an opt-in policy. Possibly the most famous nudge, certainly the most eye-catching.
 

Tags : Defaults / Organ donation / opt-in opt-out 

Source: Johnson & Goldstein (2003), ‘Do Defaults Save Lives?', Science, Vol. 302


2. 
Nudge: The authors sought to prime honesty by asking people to sign at the start of a form rather than the end when reporting how many miles they had driven on their car for insurance purposes.

In this case there was a financial incentive to report less miles driven since reporting more would mean you would pay more (i.e. a higher number in the graph implies more honesty). The results indicated the treatment to be effective at inducing more honest declarations.

Tags : priming / salience / signature / honesty 

Source: Shu et al. (2012),‘Signing at the beginning makes ethics salient and decreases dishonest self-reports in comparison to signing at the end', PNAS vol. 109.  

Amendment 19/08/2021: Please note that the findings of this paper have been called into significant questions by recent work, including a recent replication attempt finding no evidence for this effect. And also strong evidence that the field experiment was invalid due to data tampering. (LD) 

3.
Nudge: General Electric wanted its employees to stop smoking. They submitted to a Randomized Control Trial where the treatment group received cash incentives to quit. The control group received no incentives. Quitting for 6 months earned you $250, quitting for 12 months $400.

The treatment group had 3 times the success rate of the control (14.7% gave up smoking vs 5%), even after financial incentives were discontinued after 12 months (9.4% vs 3.6%). On the strength of this finding, GE now does this for their 152,000 employees.

Tags : Smoking / financial incentives / health 

Source: Volpp et al. (2009), ‘A randomized, controlled trial of financial incentives for smoking cessation', The New England Journal of Medicine.


4. 
Nudge: The ‘Save More Tomorrow’ [SMarT] program used defaults to increase employees’ savings rates by automatically increasing the % of their wage devoted to saving. Average saving rates for SMarT program participants increased from 3.5% to 13.6% over the course of 40 months. This is one of the most famous nudges.  

Tags : defaults / saving / save more tomorrow  

Source: Benartzi & Thaler (2004), ‘Save More Tomorrow’, Journal of Political Economy.


5. 
Nudge: The Behavioral Insights Team in the U.K. used social normative messages to increase tax compliance in 2011. The control group received standard tax letters. The treatment groups received the letters with an added normative messages. The difference in compliance rates between the control and the most effective treatment group was 15 percentage points.


Tags : social norms  / tax compliance / Behavioural Insights Team 

Source: Behavioural Insights Team Annual Update 2010-11. 


6.  
Nudge: A Randomized Controlled Trial using social normative messages to increase publican license renewals in Ireland. Similar to the work of the BIT, the control group received a standard renewal letter. The treatment group received a simplified letter which included a social normative message. After 1 year the treatment group had 35.5% renewal rates, the control had 29.4%. 

Tags : social norms  / compliance / pub licenses 

Source: Walsh, K. [2012, unpublished] 


7.
Nudge: Using personalized text messages for collecting fines, as per the picture. This is a very nice use of salience. The most effective treatment in this case included using the person’s name and specifying how much they owed  compared to a control of receiving a letter notifying them that they owed money.

Personalization in policy is an interesting area. Cass Sunstein, co-author of 'Nudge' believes that "personalized defaults are the way of the future". Sunstein says "it is increasingly possible for private and public institutions to produce highly personalized default rules, which reduce the problems with one-size-fits-all defaults. In principle, personalized default rules could be designed for every individual in the relevant population."
Tags : salience / tax compliance / Behavioural Insights Team 

Source: Behavioral Insights Team, ‘Test, Learn, Adapt’.


8.
Nudge: A Randomized Controlled Trial that used lotteries to encourage weight loss. Very well designed in terms of behavioral principles.

In the treatment group, a person steps on the scales in the morning. The subject’s weight is calculated & sent to a server comparing it to his weight-loss goals. The subject receives an email about his weight and a text message regarding the daily lottery. Success in meeting weight-goals enters the subject into a daily lottery with a 20% chance of winning $10 & a 1% chance of $100. If one doesn’t meet their goal, they are still entered in the lottery. But if they win, they are notified by text what they could have won, a clever use of regret aversion as motivational tool.

Tags : weight loss / regret aversion / salience / lotteries / health  

Source: Volpp et al. (2008), ‘Financial Incentive–Based Approaches for Weight Loss: A Randomized Trial’, The Journal of the American Medical Association. 


9.
Nudge: Encouraging stroke victims to take their warfarin pills via lotteries. Designed in reactance to the serious issue of patients not taking their medicine correctly.

The treatment group had a 1% chance of winning $100 contingent upon taking their pills correctly.  The control group did not have any incentives to take their pills correctly beyond the threat of dying. Adherance in the treatment group was almost 100% compared to 80% in the control [though with the caveat that the sample size is very small, n=10] . Findings like this support the argument that more should be spent on behavioral research in health-care. Dr. David Halpern of the Behavioral Insights Team has noted that in Lord Darzi’s 2008 report on the NHS  behavioral research currently receives less than 0.5% of medical research spending in the U.K.

Tags : salience / health / lotteries / adherance / financial incentive

Source: Volpp et al. (2008), ‘A test of financial incentives to improve warfarin adherence, BMC Health Services Research.


10.  
Nudge: The Behavioural Insights Team ran a 6 month Randomized Controlled Trial with Jobcentre Plus in Essex to test the impact of several changes to the way the centre operates, specifically through commitment devices & emphasis on building psychological resilence. They found job seekers in the treatment group 15-20% more likely to be off benefits 13 weeks after signing on.

Tags : job-seekers / commitment / behavioral insights team 

Source: Here is a blog post detailing it


11. 
Nudge: The American electricity company Opower use social normative messages comparing a person’s electricity usage to that of their neighbours to reduce electricity consumption. This is an area with interesting diversity of findings. Schultz et al (2007) found the use of normative messages with injunctive emoticons effective at reducing consumption. Allcott (2011) estimates this program reduces consumption by around 2% for Opower’s customers.

Interestingly, Costa & Kahn [2010] found this kind of normative nudge may backfire with political conservatives, who may actually increase their electricity consumption in reactance to it.
 
Tags : electricity usage / social norms / Opower / energy efficiency 

Sources: Schultz et al. (2007) 'The Constructive, Destructive and Reconstructive Power of Social Norms', Psychological Science.
Costa & Kahn (2010) ‘Energy Conservation "Nudges" and Environmentalist Ideology’, NBER Working Paper.

12 comments:

Cristina Raboj said...

Great job!

Not sure if it fits very well in the 'nudge' list, but here's an interesting finding. In charitable giving, seed money are the funds received prior to eliciting donations from the public. In a field experiment, List and Lucking-Reiley [2002] found that increasing the seed money from 10% to 67% of the campaign goal produced a nearly sixfold increase in contributions. The authors posit that seed money (1) signal charity credibility and reliability and (2) stand as social proof. I noticed Wikipedia does the same when they ask for donations.

The paper is called: 'The effects of seed money and refunds on charitable giving: Experimental evidence from a university capital campaign', by List and Lucking-Reiley [2002].

Unknown said...

Thanks for the heads up :)

Pelle Guldborg Hansen said...

I can't see why 3. is a nudge - it's basically about economic incentives (it could be about nudging if it compared effect of lottery with that of payment).

4. is under-described. See our list of nudges: http://www.inudgeyou.com/financial-nudge-the-classic-example-of-save-more-tomorrow/

Nick Naumof said...

Congratulations and thanks for this wonderful list.

Unknown said...

Hey Mark,
Do you think reducing the hot-cold empathy gap (Loewenstein) could be seen as a nudge that fits in your list? I just thought about why removing trays in cafeterias (http://www.nytimes.com/2009/04/29/nyregion/29tray.html) leads to less food waste and thought that if people have a tray they base their decision about how much food to take on their visceral state of hunger before they have eaten anything. Without trays they make this decision of how much to take in steps and always with an 'updated state' of hunger.

Unknown said...

Hi Max,

I'd like to include the hot-cold empathy gaps but so far I haven't found a nudge that addresses them specifically. Did you know Loewenstein and his handsome mustache are coming to Stirling in September for a talk?
http://www.stir.ac.uk/events/calendarofevents/siresymposium/

Chris N said...

Just want to thank you for putting this together - good job!

Anonymous said...

thanks !

Mark Waser said...

Isn't the graph reversed on number 2? It looks like signing at the end caused more honesty (i.e. higher mileage reported) which conflicts with the text . . . .

Unknown said...

You're correct Mark, thanks for catching that. Updated with the correct version now.

hervebrou said...

I like nudge 2,answering and putting the amount of miles in the beginning of the form rather at the end.

Unknown said...

thanks a lot for this list