BE101x Nudge Challenge Nudge condition

Behavioral Economics in Action BE101x Nudge Challenge

  1. The background and Context in which the nudge will be taken

In addition to the background of “Topic 1: Tackling the Obesity Challenge”, there are another recent study (Basu, Yoffe, Hills & Lustig 2013) where is shown that more sugar was correlated with more diabetes. Thus the goal is try a nudge to reduce the consumption of sugar.

The nudge I designed, can take place in multiple places, in different cities and countries. We only need a public or common place, like the offices of a company, the waiting rooms of an airport, or a hospital. Everywhere you can find a coffee vending machine.

  1. Precision and observability of the desired behavioral change

In my own work we have a coffee vending machine. During the course, I observe the behavior of my coworkers. Actually a population of 50 persons, where 30 of them are daily users of the coffee vending machine.

When selecting the consumption (coffee, macchiato, cappuccino, tea..) 21 of them did not change the default amount of sugar that the coffee machine offers, and 9 of them, increase the amount of sugar.

The behavior is precise and easily observable, only two options

1.- Do nothing, and accept the default amount of sugar
2.- A decision to change the amount of sugar

In case of choosing to change, it can be done to increase or reduce the amount of sugar. However I observed (empirically) that when people decide to change the amount of sugar, always choose to increase.

BE101x Nudge Challenge Control condition
BE101x Nudge Challenge Control condition – Click to enlarge
  1. The Decision Making Process and Analysis

Identification of stages of decision making process

People wants a coffee, but do not recognize the impact of consuming sugar in a excess. It’s easy to find a coffee vending machine, also is a cheap consumption and easy to use, just choose what you want and wait.

But I identify two bottlenecks:

1.- People do not identify the risks, for their health, related to an excess of sugar.
2.- The default choice of the amount of sugar, is not an active decision.

Identification of relevant phenomenon, principle and concepts covered in the course.

The main phenomenons identified in this nudge, are

1.- The default option of the amount of sugar
2.- Lack of information (a disclaimer about sugar)
3.- “Staus quo” (always choosing the same coffee with the same amount of sugar)
4.- Some self-control problems when choosing a short, sooner reward (sugared coffee) versus a larger, later reward (reduce the risk of type-2 diabetes)

  1. The Nudge

    Description and Nature

The plan is to develop a nudge in conjunction with some persuasion technique.To reduce the consumption of sugar, I propose to do with two actuations:

Remove the default option, and make an active choice.

The vending machine is calibrated with 5 levels of sugar, and the default level is 3.

Without the nudge, first you choose the type of coffee, and after, the vending machine, only wait  three seconds before the order completion. You have to be quick if you want to change the amount of sugar.

In the nudge condition, in order to complete the order, the user would need to first choose the amount of sugar proposed by the vending machine, and then choose the type of coffee.

A persuasion technique.

For those who wants to change the amount of sugar, we can persuade them by adding “extra” information in the display. In conjunction with “less/more sugar”, we can add a “more/less healthy” or even more aggressive . . . “less/more diabetes”

BE101x Nudge Challenge Nudge condition
BE101x Nudge Challenge Nudge condition – Click to enlarge

Simplicity of the Nudge

For the user, the nudge only requires a small change in the procedure. And no extra material would be needed.

Feasibility and Scalability

The nudge would be very simple. Just reprogram the machine, and add some data in the display. It can be done without too much investment.

Also the nudge would be easy to scale, we can apply it in every place where a coffee vending machine is placed.

  1. The Experiment

Factors and levels

To test the nudge, I design a “before and after” lab experiment:

– Before : no change of the default, and no persuasion.
– After : remove the default, making an active choice, and persuasion.

To really check the sugar consumption per person, the experiment would be a within-participant design.


We will choose randomly a group of persons, and under the guise to make test (could be to do some simple math, answer 5 simple questions, solve a puzzle, etc…) we will ask to go to the lab for two days.

In the lab there will be a coffee vending machine, and we will tell them “ you can take any consumption during the test”.

The first day, the vending machine are configured with the control condition, and the next day, could be a week after, configured with the nudge condition.

Outcome variable

The outcome variable is the level (from 1 to 5) of the amount of sugar that users choose.

Data analysis and prediction

An ANOVA would be used to analyze the data and check the overall level of sugar consumption.

  1. Concluding Comments

Some variations could be done in the experiment, for example, without adding the active choice, but changing the default level of sugar, reducing it from level 3 to 2. Or, as I introduced earlier, try to test the differences (if there any) if we label the persuasion, with different levels of language (“more/less healthy”,“less/more diabetes”,“longer/shorter fife”,…)