Quaranstream - how COVID-19 restrictions influenced song features on Spotify TOP-200 charts?

Abstract

This thesis focuses on researching how restriction measures, issued during the spring of 2020 and aimed towards slowing down the spread of COVID-19, influenced Spotify’s Top 200-charts in Finland, Norway and Sweden. To cover this, three research questions were composed. First, was the music listened during March – April 2020 different compared to earlier years. Second, did the selected restriction measures have statistically significant influence over audio features of the tracks appearing on Spotify’s top 200 charts. Third, if there was a shift, were there any differences between the Nordic countries.

Tracks appearing on Spotify’s top 200 charts in selected countries and during time period of interest were then analyzed by utilizing audio feature data available through Spotify’s API and compared against historical data and between countries. Finally, an explanatory forecast model was created to estimate whether there were notable exceptions connected to restriction measures.

Results show that number of Streams for Top 200 tracks decreased by -4 % to -8,7 % between 2019 and 2020 in each of the three countries. In terms of audio features, Danceability saw minor, 1,2 % to 3,8 % increases, Energy decreased by -0,3 % to -2,3 % while Tempo decreased in Norway and Sweden by -1,2% and -0,9 % consecutively and increased in Finland by 0,2 %. Valence, the feature that conveys the positivity of the music, increased notably, 7,3 % to 12,4 %.

The results suggest that listeners in each of the three countries listened to more positive sounding music during the spring of 2020. In addition, the restriction measures had a varying effect in the three countries. Most notable, in terms of statistical significance, was recommendation to shift working from home. Based on the model the restriction decreased the number of Streams and increased the Valence in each of the three countries.