Music is a world within itself, with a language we all understand (Sir Duke, Stevie Wonder)
This serious man on the left is Gustav Theodor Fechner, a German philosopher, physicist and experimental psychologist who lived between 1801 and 1887. To be honest, I don’t know almost anything of his life or work exepct one thing: he did in the 1860s a thought-provoking experiment. It seems me interesting for two important reasons: he called into question something widely established and obtained experimental data by himself.
Fechner’s experiment was simple: he presented just ten rectangles to 82 students. Then he asked each of them to choose the most pleasing one and obtained revealing discoveries I will not explain here since would cause bias in my experiment. You can find more information about the original experiment here.
I have done a project inspired in Fechner’s one that I called The Pleasing Ratio Project. Once you enter in the App, you will see two rectangles. Both of them have the same area. They only vary in their length-to-width ratios. Then you will be asked to select the one that seems you most pleasing. If you have doubts, just choose I’m not sure. You can do it as many times as you want (all responses are completely anonymous). Every game will confront a couple of ratios, which can vary from 1 to 3,5. In the Results section you will find the percentage of winning games for each ratio. The one with the highest percentage will be named officially as The Most Pleasing Ratio of the World in the future by myself.
Although my experiment is absolutely inspired in Fechner’s one, there are some differences. I can explore a bigger set of ratios doing an A/B test and I also introduce the option I’m not sure. This makes this one a bit richer.
The experiment has also some interesting technical features:
- the use of
shinydashboardpackage to arrange the App
- the use of
- to save votes in a text file
- to read it to visualize results
Will I obtain the same results as Fechner? This is a living project whose results will change over the time so you can check it regularly.
The code of the project is available in GitHub. Thanks a lot for your collaboration!
R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more…