BRIEF

Provide a service that advances positive behavior change in the world. Team must prove that a proposed behavior change idea can (or cannot) actually change behavior by running experiments to convince investors.

DELIVERED

Statistic Analysis,

Tool

Keynote, Arduino
Experiment

Post-PapaerCycle Experiment

This project is delivered based on my previous project, PaperCycle. In my previous project, I proposed an idea that we can encourage people to save more paper towels by simply teaching them how to use paper towels efficiently. The goal of this project is to run an experiment to prove that PayperCycle can change humans' behavior.

DURATION

4 Weeks

Role

Team leader, coder

Introduction

In many day to day activities there are certain behaviors that are hard wired habits: things that we do everyday without realizing that we are actually doing them. A great example of this is washing our hands after using the bathroom.

For many people, that means running your hands under water, grabbing a couple of paper towels to dry your hands off, throwing them in the trash and then going about your day. In our experiment, we sought to investigate if we can make people take just one paper towel to dry their hands, instead of a couple. Can we change this predetermined behavior by providing them with the right information at the right time?

hypothesis

People don't know how to properly use paper towels. We will significantly reduce the amount of towels used if we inform people about the techniques that make it possible to dry their hands with just one towel.

Methodology

Recruiting

All participants were unaware that they were being monitored for any period of time. The only semblance of this experiment that they could pick up was the appearance/disappearance of the signages used in the bathroom.

Set-up and Testing Method

Recording Arduino devices were installed in two of the paper towel dispensers closest to the door in the two largest gendered bathrooms, on the lower level of the school. Each Arduino setup consisted of an Arduino Uno, an Arduino Shield (for Micro SD card), a snap action micro switch, and a 2200 mAh power bank. Each unit would measure the single rotation required for the withdrawal of one paper towel within the paper towel dispenser and recorded the timestamp of the interaction. (see Fig. 1) Every towel pulled within a 3 second interval was counted towards a single user.

myProject
Based on the data, we can see nearly half of participants used only one paper towel, which is a good sign. However, there are still 56% of participants used more than one paper towels.

What if we differentiate genders...

myProject
myProject
After analysis this two pie charts, we can easily find out that the amount of male who used more than three sheets are more than the amount of females.

There might be multiple reasons behind it; maybe boys just simply dirtier than girls?
or ...?

After we placed our nudge...

myProject
myProject
myProject
There are 40% more people using just one paper towel and 30% less people using more than 3 paper towels.
Our Nudge worked!

conclusion

Throughout our sample, informational nudges could drastically decrease the amount of paper towels used in public restrooms. This will have a major impact on both financial and environmental resources.

Additional links

Experiment PaperProcess BookRaw Data Analysis

next project