The dataset for this inquiry was extracted from Spotify's Web API on November 6, 2021, and uploaded to Kaggle.com. It comprises the following:
14 songs from Taylor Swift (2006)
12 songs from Speak Now (Deluxe Edition) (2010)
19 songs from Fearless (Taylor's Version) (originally released 2008, released 2021)
6 "From the Vault" songs from Fearless (Taylor's Version) (released for the first time in 2021)
19 songs from Red (Deluxe Edition) (2012)
16 songs from 1989 (Deluxe Edition) (2014)
15 songs from reputation (2017)
18 songs from Lover (2019)
17 songs from folklore (deluxe version) (2020)
17 songs from evermore (deluxe version (2021)
In response to not owning her master recordings, Swift has been rerecording her first 6 studio albums. As of this writing (July 2022), she has released Fearless (Taylor's Version) and Red (Taylor's Version). This dataset was last updated prior to the release of Red (Taylor's Version). Thus, the data for songs from the Fearless album are related to Taylor's Version, but the songs from the Red album are the original recordings.
Alternate versions of songs–acoustic versions or pop remixes–have been removed from the dataset to limit repeat songs and ensure that the songs represented include the versions released on the standard albums.
Spotify's Web API provides several proprietary measures. Some are related to the song's composition:
Acousticness: a confidence measure from 0.0 to 1.0 of whether the track is acoustic. 1.0 represents high confidence the track is acoustic.
Instrumentalness: Predicts whether a track contains no vocals.
Speechiness: Detects the presence of spoken words in a track.
Tempo: The overall estimated tempo of a track in beats per minute (BPM).
The other measures concern the song's mood:
Valence: A measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track. Tracks with high valence sound more positive (e.g. happy, cheerful, euphoric), while tracks with low valence sound more negative (e.g. sad, depressed, angry).
Danceability: describes how suitable a track for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable.
Energy: A measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy.
Finally, there is a popularity measure which, as a percentage, represents how popular a song is according to Spotify's algorithm.
Given that Swift is most often criticized on the subject matter or feeling of her music, the mood measures were chosen as the more relevant measures and the compositional measures were not analyzed. Popularity will also be used to determine if there is a particular kind of song that Swift is most successful with.
(Definitions for measures are pulled from Spotify Developer Blog and Kaggle.com.)In addition to using Spotify’s Web API measures, I personally coded this dataset based on my knowledge of Taylor Swift's discography. The songs have been classified by their Subject, a nominal value identifying the main theme of a song. When I say these are determined in relation to the protagonist, I do not mean the speaker to necessarily be Swift. While many of these songs are at least partially-autobiographical, they are all stories with a range of creative liberities taken. They are not all in first-person, for instance. But there is usually a "protagonist," a character at the center of the story whose perspective the listener is aligned with. Additionally, when I use words relating to temporal tense, this is in reference to the timeframe of the song. For example, Swift may have written a song based on her own past romantic partner, but about the beginning of said relationship–this would be classified as a song about a relationship, not an ex.
Crush: songs about romantic interest or desire outside of an established relationship
Relationship: songs about an established romantic relationship, i.e. dating, marriage
Ex: songs about a romantic relationship that has ended, i.e. breakup songs
Self Reflection: songs which demonstrate themes of finding oneself, growing up, and self-awareness
Enemy: songs about a rival, enemy, or other negative non-romantic relationship
Family: songs having to do with family members
Other: songs which do not fit into any of the above categorizations