Perfecting Measures of Musical Preference
Where are we now?
In psychological studies, musical preference is typically measured in two ways: with the Short Test of Musical Preferences (STOMP), or with ratings of music samples. The main advantage of STOMP is that it’s, well, short. Here it is:
Researchers may favor the latter approach for its greater face validity. Instead of rating vague and broad genres, participants rate short clips of songs. Both approaches have major limitations.
Both assign a one-person-one-preference rating. Upon reflection, our listening habits are clearly not so one-dimensional. Ask yourself, what music do I listen to when I’m working? When I’m at the gym? When I’m actively listening? These may not have the same answer, and discriminating these answers may help make more accurate psychological predictions.
A problem with STOMP is that there are very important differences within genres. In “The Do Re Mi’s of Everyday Life,” pop music is grouped to the “upbeat and conventional” factor. Counterexamples abound. At first blush, “Beware the Dog” by The Griswolds sounds undeniably happy, but the lyrics are unexpectedly harrowing.
You hate so much that you’re better off alone.
Fans of this type of pop music may be noticeably different from conventional pop fans.
Enter Spotify and The Echo Nest
Spotify is a music streaming service that best distinguishes itself from its competitors with its superior understanding of musical preferences. Spotify takes note of when and what you listen to, what playlists you make, and their algorithms group these listening sessions. (e.g. discriminating “at the gym” from “actively listening,”)
Spotify owns The Echo Nest, which has an unbelievable collection of music metadata. The Echo Nest uses text and acoustic analysis to generate data for each song. (e.g. they know if a pop song has dark lyrics) One use of this data is to generate a personalized playlist called “Discover Weekly” every week, for every user.
Connecting the Dots
Spotify, as a streaming service, has access to potentially the most valuable information: what are users actually listening to.
Psychologists could partner with Spotify. Instead of asking participants to fill out a test, researchers could obtain consent to look at those participants Spotify profiles. This approach is not without its limitations. Most obviously, this kind of research would be limited to participants who actively use Spotify.
Psychologists could also explore some variables that The Echo Nest has made public:
Diversity: Measures the overall diversity of a fan’s listening by mapping the distance across the musical styles enjoyed by the listener.
Mainstreamness: Measures the overall familiarity of a user’s listening activity to determine preference for either mainstream or more obscure music.
Freshness: Measures listening habits to determine how much a user cares about new album releases vs. sticking with older music.
Adventurousness: Measures a listener’s openness to music outside their comfort zone.
References:
Rentfrow, P. J., & Gosling, S. D. (2003). The do re mi’s of everyday life: The structure and personality correlates of music preferences. Journal of Personality and Social Psychology, 84(6), 1236–1256. doi:10.1037/0022–3514.84.6.1236
http://gosling.psy.utexas.edu/wp-content/uploads/2014/09/stomp.pdf
