Skip to main content
Figure 1 | EPJ Data Science

Figure 1

From: Prediction of employment and unemployment rates from Twitter daily rhythms in the US

Figure 1

Population-weighted Pearson correlation of employment and unemployment levels with hourly activities. Errorbars are calculated using bootstrapping \(n=1\mbox{,}000\) times. The hours between 6am and 8pm correlate significantly positively with employment and negatively with unemployment. This relationship turns out to be exactly the opposite during the night. Regarding employment, the most distinguishing hours are 1am (most negative correlation) and 12pm (most positive correlation).

Back to article page