About the Data

(GitHub) Get the Data

Spotify Web API Data

For track audio features such as duration, valence, etc. I utilized the Spotify Web API. Paired with Burati's compilation of the full Hot 100 history, I collected the audio features data for every track and added it to the information present in Burati's data, including some features not in this project such as "danceability" and "acousticness". It is important to take this algorithmically-generated data with a grain of salt when it comes to determining the values for these features. For example, Tchaikovsky's "Dance of the Sugarplum Fairy" has a danceability of only 0.323, yet it is a famous ballet piece. The algorithm may have narrow interpretations of some characteristics, and checking the documentation is important for interpreting the data. The data is available in the GitHub link above.

---

Race, Gender, and Collaboration Data

For each of the five decades in my dataset, I randomly selected 30 tracks. If there was too much uncertainty about an artist's identified gender or race, another track was randomly selected in order to maintain as much accuracy as possible. If there are any discrepancies in the dataset, I am happy to correct that in this project if you submit an issue on the GitHub page linked above.