Presenter Information

Anthony CampanaFollow

Loading...

Media is loading
 

Major

Business Administration

Anticipated Graduation Year

2022

Access Type

Open Access

Abstract

The Billboard Hot 100 has sought to track the music preference of Americans since 1958 by compiling a list of the 100 most listened songs in a given week. Spotify, a music streaming company, offers an interesting dimension to this index by way of their API. Each song in the database is tagged with numerical attributes including danceability, valence (happiness), energy, key, loudness, mode, speechiness, acousticness, instrumentalness, and liveness. Each attribute is determined using algorithms creating by social scientists focused on Natural Language Processing (NLP), tempo, and pitch. This study seeks to assess the relationship between these attributes in the Billboard Hot 100 and changes major economic indicators including Real GDP, the Unemployment Rate, and the Consumer Pricing Index (CPI). The dataset was then run through a Decision Tree model with the intention of assessing the relationship between the variables. The model found that the change in Real GDP was the strongest predictor of how music preferences would shift. Of the attributes, valence (happiness) was the most affected. This information reveals an interesting relationship between the economy and major economic indicators. Further analysis could be beneficial to record labels when deciding which artists to pursue for the upcoming year.

Faculty Mentors & Instructors

Professor Kmet

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.

Share

COinS
 

Assessing the Relationship Between Economic Indicators and Music Preference in the US

The Billboard Hot 100 has sought to track the music preference of Americans since 1958 by compiling a list of the 100 most listened songs in a given week. Spotify, a music streaming company, offers an interesting dimension to this index by way of their API. Each song in the database is tagged with numerical attributes including danceability, valence (happiness), energy, key, loudness, mode, speechiness, acousticness, instrumentalness, and liveness. Each attribute is determined using algorithms creating by social scientists focused on Natural Language Processing (NLP), tempo, and pitch. This study seeks to assess the relationship between these attributes in the Billboard Hot 100 and changes major economic indicators including Real GDP, the Unemployment Rate, and the Consumer Pricing Index (CPI). The dataset was then run through a Decision Tree model with the intention of assessing the relationship between the variables. The model found that the change in Real GDP was the strongest predictor of how music preferences would shift. Of the attributes, valence (happiness) was the most affected. This information reveals an interesting relationship between the economy and major economic indicators. Further analysis could be beneficial to record labels when deciding which artists to pursue for the upcoming year.