This repository contains the code of the master thesis from Long Hoang Nguyen. In the course of the thesis, we explored the effects of a tidy vs. messy online environment on human behavior and choices. This exploration included an online experiment with two different prototypes for a tidy and messy website. Both created prototypes use the same frontend and backend technologies, with the difference lying in the CSS code.

• Frontend: SurveyJS and Bootstrap
• Backend: Google Sheets API and Google Apps Script

• backend_scripts: Contains a Google Apps script that allows for communication between the Frontend and Backend.
• clean_website: Contains the code for the tidy website interface.
• messy_website: Contains the code for the messy website interface.
• welcome_page: Contains the code for a welcome page that redirects the user to either experimental condition randomly.

In case you are interested in using this code, you should:

1. Git clone this repository.
2. Open the index.html file of the welcome_page and redirect to either the clean_website or messy_website.
3. Check if there are any issues with either website (e.g., Displacement of images).
4. Read the comments in the getData.gs file in the folder backend_scripts and set up your Google Sheets API access.
5. Try to collect data using either interface.

To objectively determine whether these developed prototypes are tidy or messy, we included metrics of Visual Clutter - in particular Feature Congestion and Subband Entropy. As the setup can be inconvenient on Windows machines, we set up a Google Colab notebook that can be accessed here. Keep in mind, that you probably have to configure your own Colab environment before using the code.

Feel free to use the contents in consideration of Creative Commons Attribution 4.0. In case there are any further inquiries, you can also contact the author.