🔗 Interactive Dataexploration

Explore the data on your own at: 

http://creative-area.com/feminist_music/

Intro

For many years feminists fought and are still fighting for equality and women rights. Music is one of the most universal ways to communicate ideas and express oneself, with the potential to reach a broad audience. Of course, the industry itself has it shortcomings when it comes to gender equality, diversity und inclusion. In spite of that more and more remarkable feminist artists are on the rise and have a positive impact inside this space. With datasets from Wikipedia, the world’s biggest free encyclopedia and Spotify, one of the biggest music streaming platforms in the world as primary sources, the following data visualisation takes a closer look on those artists.

Through carefully analyzing and polishing these large datasets we can begin to explore questions like: What characterises a feminist artist? Which role do parameters like popularity, location, music genre or gender play? Where are the differences and similarities? Is there a focus on a specific genre or how do they compare to the general music industry?

Please note that the following graphics won’t equate to a universal and definitive conclusion on this research matter. Some of the datasets we use are inherently biased. Still the datasets are coherent enough to warrant further inspection. The visualisation of patterns and connections inside those datasets, with the goal of facilitating discourse and exploration was paramount.

Scope and Constraints

During early days of development we had a very different kind of data visualisation in mind. Our goal was to offer a complex exploration tool with a multitude of different options and filters. The idea was to let people explore the dataset on their own, giving them as many tools as possible while refraining from displaying our own conclusions. 

After additional research and a more concrete understanding of our datasets we decided against our initial concept. In hindsight there were two deciding factors: 1. creating such an expansive tool takes a lot of time and can't easily be scaled up or down, potentially leaving us with an unfinished tool at the end of the course. 2. the more we learned about our dataset and the subject matter, the more we realised that there are many interesting stories to be told and comparisons to be made. Simply delivering an exploration tool without shining a light on those findings just wasn't an option for us anymore.

Our new approach kept a clear focus on data visualisation but gave opportunities to create a narrative and highlight interesting facts. Working in segments also gave us great freedom to scale the project. Adding or removing different elements depending on time constraints and content value.

🔗 Playlist: Feminist Musicians

As the data-exploration focuses on the musicians and comparable data of their music, we felt the need to add a possibility to get a feel for their music and lyrics. Of course the only option is a 165 hour long playlist of the top 10 songs of every feminist artist we found on spotify.

Listen to the playlist:

⚧ Spotify Playlist: Feminist Music

Please feel free to recreate the playlist on other streaming services and contact us in order to link it here.

Gender Distribution

To take a look at gender distribution between feminist artists and the rest of the music industry we started by comparing our data from Wikipedia.com with results from the “Inclusion in the Recording Studio?”-Study held by the USC Annenberg Inclusion Initiative in 2019. The study analysed the “Top 100 End-Year Billboard Charts 2012-2018” and gives insights into the gender distribution in popular music. The comparison made abundantly clear that there is a big divide in term of gender distribution between our two data sets. The Wikipedia-List of feminist artists is almost 100% comprised of female or mixed-gender artists, with some rare male exceptions like Lemmy or John Legend. Meanwhile the study revealed a very different picture with almost 80% of artists in the Top 100 between 2012-2018 being male.

While Spotify sadly doesn’t provide clear data on the exact gender distribution of artists inside the streaming service, they offer a lot of informations about their streams. Taking a closer look at those daily updated streaming numbers (reference date: 23.01.2020) reveals further prove of an less than equal playing field. According to the statistics the service has 43,6% female or mixed-gender listeners but only 22,4% of streamed music by female artists. This rather disappointing result can be easily attributed to a specific cause or to a general bias on Spotify’s behalf. Listeners using Spotify’s editorial Playlist actually stream 24,8%, 2,4% higher than the average, while people that listen to their own playlists are even below-average with 20,7% of streamed female artists. Listening to Spotify’s algorithmically personalized Discover Weekly playlists, leads to the worst results, streaming only 16,9% from female artists.

Chronological View

Our data shows a steady increases of feminist artists over the decades. While the dataset is inherently focused on north America, starting in the 50s with hotspots on the west and east coast of the USA, a more diverse distribution all over the world can still be observed during the 70 years till 2020.

Artist List

In order to allow for additional personal research, there are references to the raw artist data within many data visualizations. The full list of artists and all of their data we managed to get from wikipedia and spotify is available at any point on the right side of the window. 

From there, it is possible to get to the corresponding wikipedia article, their website if listed or listen to the artist via spotify.

Artist Popularity

Visibility in front of a global audience gives an artist the possibility to make their opinions and view have an impact on culture and society. In order to explore the reach of feminist musicians and their visual variety, we chose to list all artists by their popularity with their latest images according to spotify.

Representation on Spotify

Being represented on Spotify means access to a global audience. Being represented on Wikipedia is in itself a sign for popularity. „Being popular“ has a rather murky definition, so comparing the two metrics can serve as a first step in narrowing down a quantifiable conclusion. Our data shows, that about two thirds of all feminist artists from our wikipedia list also can be found on Spotify, thus expanding their reach substantially.

Popularity & Followers

While our data-set shows rather positive results in terms of popularity, there are additional factors that can’t be effectively displayed in this visualization but are important to note: f.e. Some artists are sympathetic to the feminist movement but are reluctant to identify as feminist in fear of compromising their artistic career and being reduced on the topic of feminism, others never state they are feminists but become a symbol for feminism by being successful women in a male dominated industry.

Popularity Score

Quiet a few feminist artists on our list have very high 90+ ratings in popularity, showing that artists with feminist ideas and thoughtful songs can gain mass appeal and huge followings.

Followers

While popularity is measured through a combination of monthly listeners, followers and other factors, focusing on followers only, offers a more streamlined perspective and shows bigger gaps in terms of reach.

Top 25 Artists by Followers

Looking at the top 25 of artists with the most followers on Spotify, some feminist artists like Ariana Grande, Taylor Swift and Beyoncé reach some of the top spots and have up to 41.3 million followers.

Genres

We counted over 250+ different entries, revealing a very diverse and artistically rich spectrum of musical genres. The most prominent genre is Pop with it’s many different variations. Other noteworthy genres include „Lilith“, which is comprised of music played at the Lilith Fair, a concert tour, founded by Sara McLachlan in 1997. The financially successful tour, exclusive to female solo-artists and female-led bands was created in response of concert promoters repeatedly refusing to book two female artists in a row. Another interesting genre is Spotify’s own algorithmic genre „Escape Room“, that is part of a growing list of new micro-genres to better categories and cater to listeners music tastes. As such, it can’t be pinned down to a clear direction but rather entails a wide variety of musical impressions.

Instruments

When choosing instruments, feminist artists seem to have three clear favorites. “Vocals, guitars and pianos” with most opting for vocals. Beside those three rather classic choices, our findings again show a wide variety of different instruments. With outliers like the oboe or the laserharp. It’s important to note that this statistic in particular is to take with a grain of salt, considering many of our artists play more than one instrument and no reliable way for us to give each instrument a more accurate emphasis.

Further Thoughts

Additional informations about our data

Wikipedia offers the biggest, free encyclopedia on the internet, with almost 6 million unique articles on its english speaking „wikipedia.com“-Domain alone. This made Wikipedia our first choice while looking for a list of feminist music artists that was as complete as possible. During our research we faced a few defining problems that would taint our data in different ways. One of the biggest decisions was the choice of which version of wikipedia we should use. Choosing the english version would focus mostly on american and english speaking artists while also including more famous artists from other countries. Choosing to combine lists of different wikipedia versions around the world would be difficult from a linguistic standpoint. Not being able to speak most of the languages fluently and no real way to verify the datasets, resulting in a skewed list with no tangible parameter other than “we are able to speak that language”. So we settled for the english wikipedia version utelising a list of “feminist musicians”, that missed some of the bigger and well-known feminist artists but served as a good starting point. We incorporated additional relevant Wikipedia-Lists and began to expand the list with artists from around the world, that we found during our research, in case they also had an entry on Wikipedia.

Like previously stated, Spotify is one of the biggest music-streaming services in the world, with 248 million active users per month (Q3 2019). Of course we are well aware that Spotify isn’t without its flaws and it’s possible that influential feminist artists aren’t present on Spotify simply because they don’t agree with some of its business practices. Nevertheless, after examining the sheer size and variance of its library we deemed it a good, comparable counterpart to our Wikipedia data set. While both aren’t ideal, each offering a good amount of usable data, that operate on similar parameters regarding relevance and popularity.

Refining the dataset

Working with Wikipedia turned out to be a bigger struggle than anticipated: With its biggest strength, the option to create and edit articles by everybody also being one of the biggest problems for us. With only lose rules regarding formation of data and very lax policies in terms of curating articles, especially on the english version of wikipedia. This led to a lot of problems, while trying to clean up and structurize the datasets. Coming up with a viable solution to extract a useable list out of this mess of scrambled info-bits expanded the scope of this project immensely. In the end we managed to unscramble about 96% of the data, leaving us with 4% of data that would have to be analysed manually. With the alternative being to comb through 15875 lines of code and in light of very tight time constraints, we decided to ignore the occasional data-hiccup(f.e. missing links or redundant commata).

Outlook

We had a lot of different ideas for additional comparisons and visualizations that we sadly couldn't incorporate into the project mostly because of time constraints. Here are some of the ideas that would be thinkable for a potential expansion of this project: Connect the dataset with twitter (wordclouds, relevant tweets, popular topics among feminist artists, connections between artists, reach, popularity, etc.), revisiting our initial idea of a free exploration tool and to organically implement it into the project, delve deeper into artist connections inside of Spotify, highlight more artists of different decades and expand the timeline to 1900 or even further, use more datasets and expand existing datasets.

Get the raw data

Transparency is critical to make a contribution in research overall and specifically on a topic such as feminist music.

Therefore we encourage anyone to build upon our intense work on the dataset and use it to explore new aspects and generate new visualizations. 

The raw artists data, that is used for all data visualizations, is available to download as .csv and .json (suggested, as better structured) on:

http://creative-area.com/feminist_music/

The data of Wikipedia and Spotify was accessed January 2020.

Many thanks to the Prof. Franziska Morlok, Dr. Julia Meer and the students of the 2019/20 class „Sichtbarkeit gestalten“ within the University of Applied Sciences Potsdam, Germany.

Sources

Wikipedia Category: Feminist musicians, last visit: 23.01.2020, 13:41
https://en.wikipedia.org/wiki/Category:Feminist_musicians 

Wikipedia Category: Feminist rappers, last visit: 23.01.2020, 13:44
https://en.wikipedia.org/wiki/Category:Feminist_rappers

Wikipedia: Women’s Music, last visit: 23.01.2020, 13:25
https://en.wikipedia.org/wiki/Women%27s_music

Wikipedia: List of most-streamed artists on Spotify, last visit: 23.01.2020, 13:32
https://en.wikipedia.org/wiki/List_of_most-streamed_artists_on_Spotify

Study: Inclusion in the Recording Studio?, USC Annenberg Inclusion Initiative, last visit: 21.01.2020, 20:12
http://assets.uscannenberg.org/docs/aii-inclusion-recording-studio-2019.pdf

Spotify Listening Patterns by Gender, last visit: 20.01.2020, 17:59
http://everynoise.com/gender_tldr.html

25 Top Feminist Anthems, last visit: 18.01.2020, 19:10
https://www.billboard.com/articles/news/pride/8022687/top-feminist-anthems-songs

17 Feminist Songs That Were Ahead Of Their Time, last visit: 18.01.2020, 17:04
https://www.huffpost.com/entry/17-feminist-songs-that-were-ahead-of-their-time_n_56fc6b46e4b0daf53aeeaf5a

What Is “Escape Room” And Why Is It One Of My Top Genres On Spotify?, last visit: 20.01.2020, 12:45
https://festivalpeak.com/what-is-escape-room-and-why-is-it-one-of-my-top-genres-on-spotify-a886372f003f

Lesley Gore: The Jewish Feminist Lesbian Pop Star Ahead of Her Time, last visit: 21.01.2020, 11:50
https://www.heyalma.com/lesley-gore-the-jewish-feminist-lesbian-pop-star-ahead-of-her-time/

Statista – Number of Spotify monthly active users (MAUs) worldwide from 1st quarter 2015 to 3rd quarter 2019, last visit: 21.01.2020, 20:05
https://www.statista.com/statistics/367739/spotify-global-mau/

Statista – Number of Spotify premium subscribers worldwide from 1st quarter 2015 to 3rd quarter 2019, last visit: 21.01.2020, 20:05
https://www.statista.com/statistics/244995/number-of-paying-spotify-subscribers/