Stats-Fun with ESC-BPM-Analysis and Economic Inequality

Lustige Scheinkorrelation von Ars Technica, die sich die BPM (Beats per Minute) der, ähm, Songs vom Eurovisiontrash-Contest angesehen und mit den Zahlen des Gini Index („a measure of a country’s level of inequality“) angesehen haben. Ergebnis: Länder mit einer höheren Arm/Reich-Schere und ungerechterer Wohlstandsverteilung spielen scheinbar schnellere Mucke.

Screen-Shot-2016-05-15-at-5.17.19-PM-640x632As the scatter plot [to the right] shows, there is a positive correlation between a song’s BPM (the x-axis going across) and the competing country’s Gini Index (the y-axis going up). While it isn’t the case for every country, on average the tempo of the song was faster when the level of inequality was higher in that country. Using Graph Pad’s Prism 7 statistical software, a Student’s T-Test showed that this was a highly significant result with a p-value of 0.0001 and a t-value of 24.9838. (Scientific studies consider a statistical result to be “significant,” or less likely to be the result of a random error, if the p-value is below 0.05. Likewise, the further a t-value is from zero, the more likely it is that you can reject the “null” hypothesis that there is no significant difference.) In other words, this analysis shows that there is a strong positive correlation that is highly statistically significant […]

Economist Noah Smith looked at our statistical analysis and offered a couple of possibilities via e-mail: „Real explanation: Probably an omitted variable. South Europe probably has faster songs traditionally, and also happens to be more unequal. Fun explanation: Upbeat songs distract the proletariat from the evils of late stage capitalism and create false consciousness!“

Vorher auf NC:
Scheinkorrelationen
Spurious Correlations
ESC BPM Analysis: 128 Beats Per Minute suck big time