There’s a school of thought that says if you break anything down into the correct number of data points, you can understand how it works and the effect this thing has.
Take music, for example. Is there any way to analyze a piece of music in such a way that tells us why it makes us feel the way it does? For example, why do certain songs make us want to dance?
New research from Columbia School of Business and INSEAD, a French business school, says they’ve got it figured out.
Using chart data from Billboard going back to 1958 and streaming info from Echonest (the data-crunching arm of Spotify), the research analyzed things like beat, instrumentation, key signature, and valence to create an algorithm to determine the danceability of a given track. It also looked at things like consistency of beat, meaning that it favoured songs that allowed dancing at a steady rate from beginning to end. In other words, if the song featured a break or bridge that interrupted the beat, it was ranked lower.
The result was a list they called “The Most Danceable US Billboard Number Ones since 1958.”
- Give it to Me, Timbaland (2007)
- SexyBack, Justin Timberlake (2006)
- Hot in Herrem Nelly (2002)
- Ice Ice Baby, Vanilla Ice (1990)
- Pop Muzik, M (1979)
- Another One Bites the Dust, Queen (1980)
- Funkytown, Lipps, Inc. (1980)
- Can’t Nobody Hold Me Down, Puff Daddy (1997)
- Baby Got Back, Sir Mix-A-Lot (1992)
- Billie Jean, Michael Jackson (1983)
- Bad Girls, Donna Summer (1979)
- I’ll Be Missing You, Puff Daddy & Faith Evans (1997)
- Hollaback Girl, Gwen Stefani (2005)
- Fancy, Iggy Azalea (2014)
- Then Came You, Dionne Warwick (1974)
- Boogie Fever, Sylvers (1976)
- Low, Flo Rida (2008)
- Hypnotize, Notorious B.I.G. (1997)
- It’s Still Rock and Roll to Me, Billy Joel (1980)
- In Da Club, 50 Cent (2003)
You can bet that songwriters will want to use this list to help predict what new songs could be danceable hits. Meanwhile, you might want to dance along to this playlist as you read the original research paper.