Introduction

Pop music, or popular music, is fairly recent term and originated in the 1950s. This genre of music is most likely the type you would hear on the radio, and most artists in this genre have songs in the top 100. The type of music produced in this genre changed over time. Around the 1990s female singing groups became more and more popular. In 1996, Spice Girls became one of the most popular and well-known groups. From the 90s-present more and more females are rising to the top in the music industry. With this in mind, I am interested to see if the role women play in music has changed over time.

Hypothesis

Has the role women play in music change over time? Are the popular positive and negative words in each album the same over time for female pop artists? Lastly, is there a central focus or theme in the selected albums examined?

Methodology

For the 3 decades examined, I was able to find the top male and female pop artists of that time period. The 1980s, 1990s and 2000s/2010s were examined. For each decade, a random number generator was used to select an artist. Because females were only being analyzed for this project, if the number landed on a male artist or a male artist group, I randomly generated another number until landing on a female artist or group. If a number was not assigned to a pop artists already (this was the case for the 1990s pop artists), I assigned each artist a given number before using the random number generator to select which female artist to analyze for each decade. After this randomization process was completed for each decade, I selected the top album for each decade. 2000s and 2010s was combined because these decades shared close to the same types of pop artists. Data was imported into R using the Genius API. Each album was separated and then analyezed to find the most common words in each album using the code as follows:

# lemonade <- genius_album(artist = "Beyonce", album = "Lemonade")

# lemonade %>% 
#  unnest_tokens(word, lyric) %>% 
#  anti_join(stop_words) %>% 
#  count(word, sort=TRUE) -> lemonadeCount

I created a table of what the leomadeCount variable looks like below:

Word Count
love 82
slay 49
daddy 27
freedom 21
feel 19
night 18
money 17
catch 16
shoot 16
baby 15

Next, the Bing lexicon in R was used to create the sentiment analysis. Bing categorizes each word, in this case the most popular words in the album, as either negative or positive. Bing and the function innerjoin were used to merge tables. I decided to use Bing instead of either Afinn or nrc. Bing tends to remove more words because it does not have a range. One downside to using bing is that it is a boolean variable, meaning it will place words in either a positive or negative pile. As we know, words can be described in ways other than positive and negative. The coding process was a two-step process. First, functions were used to further filter the data and then a graph was created. The mutate funtion was passed through the pipe operator to order the words from most used to least used. That’s why the most popular positive and negative words were shown at the top of the graph. The facet_wrap option reflects our use of one categorical variable, ‘sentiment’. Next, the coord_flip() option switches the x and y axis so that words are reflected on the left hand side, making the chart more clear. Because the coord_flip() layer was used first, the xlab actually reflects the y axis and vice versa. Additionally, extra layers were added in the code that change the colors of the outline and bars and add titles and axises. This process was done for the top 20 positive and negative words for each artist’s top album. Here is the code:

# lemonadeSentiment <- lemonadeCount %>% 
#     inner_join(get_sentiments("bing")) %>% 
#     ungroup()

# lemonadeSentiment %>% filter(sentiment=="positive") -> positiveNew
#     positiveNew %>% head(20) %>%
#     mutate(word = reorder(word, n)) %>%
#     ggplot(aes(word,n, fill = sentiment)) +
#     geom_col(show.legend = FALSE, color="black", fill="green") +
#     facet_wrap(~sentiment, scales = "free_y") +
#     coord_flip() +
#     xlab("Word") +
#     ylab("Count") +
#     ggtitle("Lemonade Positive Sentiment Analysis") +
#     theme(plot.title = element_text(hjust = 0.7))

# lemonadeSentiment %>% filter(sentiment=="negative") -> negativeNew
#     negativeNew %>% head(20) %>%
#     mutate(word = reorder(word, n)) %>%
#     ggplot(aes(word,n, fill = sentiment)) +
#     geom_col(show.legend = FALSE, color="black", fill="red") +
#     facet_wrap(~sentiment, scales = "free_y") +
#     coord_flip() +
#     xlab("Word") +
#     ylab("Count") +
#     ggtitle("Lemonade Negative Sentiment Analysis") +
#     theme(plot.title = element_text(hjust = 0.7))

2010/2000s: Beyonce

Beyonce has been an Icon in Pop Culture for years. Known as, “Queen B”, Beyonce started her career as a member in the female R&B group, Destiny’s Child formed in 1997. In 2005, the group split. She released the album Dangerously in Love soon after. To date, Beyonce’s top album is Lemonade (2017). When this album was released, it had up to 9 Grammy Awards. In her career, she has sold over 100 million records worldwide. She has also won 22 Grammy Awards.

1990s: Spice Girls

In 1994, the Spice girls became a female pop artist group in London, England. Virgin records signed their release and in 1996 they debued one of their top songs, “Wannabe”. The Spice Girls’ top album was Spice (1996). As mentioned in the introduction, the Spice Girls were one of the first female singing groups to exist. Their popularity influenced the rise of female singing groups today.

Here is a count of the most popular words before the sentiment analysis.

Word Count
wanna 61
dance 60
love 54
time 36
baby 34
lover 33
naked 29
mama 22
yeah 21
gotta 20

1980s: Madonna

In 1981, Madonna rose to fame when she went to Gotham Records to get her singing career started. Within a year, she had her first number one hit titled “Everybody”. Madonna’s top album was Like A Prayer (1989). Her nickname was the “Queen of Pop” and also an icon in the world of music. She is most popular because she became a sensation in a male-dominated 80s music scene. Interestingly enough, some say she is a “barometer of culture that directs the attention to cultural shifts, struggles and changes.”

Here is a count of the most popular words before the sentiment analysis.

Word Count
love 64
cherish 27
time 27
prayer 24
baby 21
anymore 15
eyes 14
joy 14
voice 14
dream 13

Word Clouds

Word clouds were generated using the wordcloud2 package in R. Options were used that allowed the color of words, font, and background color to be changed. The word clouds reflect the most frequent words in each album. The more frequent the word, the larger it will appear. The following code and word clouds are below:

# library(wordcloud2)
# wordcloud2(lemonadeCount, color = "random-light", 
#         backgroundColor = "black", fontFamily =  "Helvetica")
A word cloud of Beyonce’s most popular words in her album, Lemonade (2017)

A word cloud of Beyonce’s most popular words in her album, Lemonade (2017)

Spice Girls word cloud of Spice (1996) album

Spice Girls word cloud of Spice (1996) album

Word cloud of Madonna’s album, Like A Prayer (1989)

Word cloud of Madonna’s album, Like A Prayer (1989)

Results

The results were interesting among the artists examined. It was interesting which words the bing lexicon identified as positive and negative. Most I agree with, but some are questionable. In Beyonce’s negative sentiment analysis, the word fall was one of the top 20 negative words in Lemonade. When I saw this, I initially thought of the term “fall in love” which is interpreted positively. This goes to show how hard it is for the computer to categorize words under the umbrella of positive or negative without additional context.

For all three females, love was the highest counted positive sentiment in each top album. The frequency of positive words surpassed the frequency of negative words by a milestone among all three artists. The most popular positive words appeared about 55-80 times while the most popular negative words only appeared 7.5-15 times throughout all three albums. This lead the ratio of positive words to negative words to be very high.

The count of words reflected a more positive tone in general before looking at the sentiment analysis. In Madonna’s case, there seemed to be a larger gap between the first and second most frequent words. Love was counted 62 times in the album and cherish was counted 27 times. Beyonce had a similar pattern, but the Spice Girls had a smaller range of counts.

Conclusions/Future Research

Based off of the results, it is hard to conclude if the role women play in music has changed over time. The frequency tables showed in general the pop artists tend to sing more about topics and themes that the bing lexicon counts as more positive. However, when comparing the frequency of words among the three artists it is hard to make a definite conclusion. Words such as love, trust, joy, grace, strong, faith, and promise seemed to be used in songs across the board. These words resemble and mirror words of empowerment. Because female pop artists tend to attract a mostly teenage/early adult females, I would like to think that empowering women has been a general trend since at least the 1980s.

Love was a common theme and the top word used across all albums for female pop artists. This makes sense because the nature of pop music seems to attract a younger audience in general. Additionally, most song writers write about experiences they’ve had and finding love or losing love tends to be very impactful. Above else, love trumps all! It’s great that this is reinforced in mainstream music.

In order to construct a more in depth analysis, male pop artists need to be examined. It would be interesting to see if male pop artists have similar or different most popular positive and negative words in their albums. For women, love seemed to be a common positive word and theme across the artists examined. However, would love be the top popular, positive word for males? Would the top positive words vary amongst males, while they were similar for females? This being said, there is a lot more to discover about this topic and I can’t wait to explore more about it for my final project.