Butterflies Group - Megan Sierra and Mallory Walker
 |
 |
In this project, we would like to explore how music has changed over time. From timeless classics to era-defining hits and contemporary tracks, by analyzing listening trends and genres, we can reveal how the world’s mood and environment may have shaped the sounds of its time.
The purpose of this project is to explore how each generation and the world’s climate may have shaped the sounds of its time.
Our data set (“Spotify Global Music Dataset (2009–2025), authored by Warda Bilal) was found on Kaggle (Spotify Dataset), which contains 8,583 records and 15 variables offering comprehensive information about songs and artists.
library(tidyverse)
library(knitr)
# Variable table
data.frame(Variable_Names = names(spotify)) %>%
knitr::kable(
caption = "Variable Names in Spotify Dataset"
)| Variable_Names |
|---|
| track_id |
| track_name |
| track_number |
| track_popularity |
| explicit |
| artist_name |
| artist_popularity |
| artist_followers |
| artist_genres |
| album_id |
| album_name |
| album_release_date |
| album_total_tracks |
| album_type |
| track_duration_min |
Out of the 15 variables, we chose to focus on these six variables:
album_release_date
track_name
artist_name
track_popularity
artist_genre
artist_name
You can interact with the data using the search box, such as limiting results to Bruno Mars or a chosen year.
#install.packages("DT")
library(DT)
datatable(spotify)This project analyzes the data across two levels:
Trend in popularity among songs across generations
Artists that have released the most songs from 2000-2025
Generation X is the generation following behind the Baby Boomers. Generation X is more commonly linked with skepticism, independence, and a tendency to question institutions. One of the clearest ways this identity took shape was through music. Growing up, the sound that defined Generation X’s formative years were different types of rock, like classic and pop rock. This showed that their culture was ready for rebellion. Gen X’s parents (the Silent Generation and early Boomers) had grown up with safe, polished pop. Rock is loud, guitar-driven, sometimes provocative. It was the sound of pushing back. It felt dangerous in a way that resonated with youth.
The most popular song for the Gen X time period was The Chain by Fleetwood Mac, which was made in 1977. This song resonated with Gen X because it defined independence and and the importance of loyalty. The most popular song of this generation, The Chain, reflects the generational emotional and social themes of Gen X .
top10_songs_1965 <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 1965 & album_release_year <= 1980,
!is.na(track_popularity)) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
slice_max(track_popularity, n = 10, with_ties = FALSE) |>
ggplot(aes(x = reorder(paste(track_name, "-", artist_name), track_popularity),
y = track_popularity,
fill = factor(album_release_year))) +
geom_col() +
coord_flip() +
scale_fill_brewer(palette = "RdPu") +
scale_y_continuous(limits = c(0, 100), breaks = seq(0, 100, by = 10)) +
labs(
title = "Top 10 most popular songs (1965-1980)",
x = "Track — Artist",
y = "Popularity score",
fill = "Year"
) +
theme_minimal() +
theme(legend.position = "right")
top10_songs_1965
This line chart shows the trend in the most popular song for every year between 1965-1980 (Gen X).
library(plotly)
genX_data <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 1965 & album_release_year <= 1980,
!is.na(track_popularity))
genX_top_song <- genX_data |>
group_by(album_release_year) |>
slice_max(track_popularity, n = 1, with_ties = FALSE) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
ungroup() |>
arrange(album_release_year)
plot_ly(
data = genX_top_song,
x = ~album_release_year,
y = ~track_popularity,
type = "scatter",
mode = "lines+markers",
text = ~paste(
"Year:", album_release_year,
"<br>Popularity:", track_popularity,
"<br>Top Song:", track_name,
"<br>Artist:", artist_name
),
hoverinfo = "text"
) |>
layout(
title = "Top song popularity by year (Gen X: 1965-1980)",
xaxis = list(title = "Year", dtick = 1),
yaxis = list(title = "Track popularity", range = c(0, 100))
)Millennials are the generation following Generation X. Millennials are often linked with adaptability, diversity, optimism, and growing up during rapid technological change. One of the clearest ways this identity took shape was through music. Growing up, the sound that defined Millennials’ formative years included pop, hip-hop, R&B, alternative rock, and electronic music. This showed that their culture embraced variety and constant change. Unlike previous generations shaped mainly by one dominant genre, Millennials were raised during the rise of MTV, digital downloads, and the internet, where many styles existed at once. Their music was catchy, emotional, experimental, and globally influenced. It reflected a generation learning to navigate a fast-moving, connected world.
The Millenial generation was the largest generation in Western History. Millenials saw the rise of the Internet and describes as the most globalized generation. The most popular song for the Millenial Generation was Every Breath You Take by The Police which was made in 1983. This song has themes of obsession and surveillance which relates to the rise of technology with this generation.
top10_songs_1981 <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 1981 & album_release_year <= 1996,
!is.na(track_popularity)) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
slice_max(track_popularity, n = 10, with_ties = FALSE) |>
ggplot(aes(x = reorder(paste(track_name, "-", artist_name), track_popularity),
y = track_popularity,
fill = factor(album_release_year))) +
geom_col() +
coord_flip() +
scale_fill_brewer(palette = "RdPu") +
scale_y_continuous(limits = c(0, 100), breaks = seq(0, 100, by = 10)) +
labs(
title = "Top 10 most popular songs (1981-1996)",
x = "Track — Artist",
y = "Popularity score",
fill = "Year"
) +
theme_minimal() +
theme(legend.position = "right")
top10_songs_1981
This line chart shows the trend in the most popular song for every year between 1981-1996 (Millennial).
library(plotly)
Millen_data <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 1981 & album_release_year <= 1996,
!is.na(track_popularity))
Millen_top_song <- Millen_data |>
group_by(album_release_year) |>
slice_max(track_popularity, n = 1, with_ties = FALSE) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
ungroup() |>
arrange(album_release_year)
plot_ly(
data = Millen_top_song,
x = ~album_release_year,
y = ~track_popularity,
type = "scatter",
mode = "lines+markers",
text = ~paste(
"Year:", album_release_year,
"<br>Popularity:", track_popularity,
"<br>Top Song:", track_name,
"<br>Artist:", artist_name
),
hoverinfo = "text"
) |>
layout(
title = "Top song popularity by year (Gen Millenial: 1981-1996)",
xaxis = list(title = "Year", dtick = 1),
yaxis = list(title = "Track popularity", range = c(0, 100))
)Generation Z is the generation following Millennial’s. Generation Z is more commonly linked with individuality, emotional openness, digital fluency, and a tendency to challenge social norms. One of the clearest ways this identity took shape was through music. Growing up, the sound that defined Generation Z’s formative years included alternative pop, indie rock, hip-hop, electronic music, and revived older hits from past decades. This showed that their culture valued freedom of expression and authenticity. Unlike earlier generations that relied on radio or television, Gen Z grew up with streaming and social media, where every genre was instantly available. Their music was emotional, experimental, nostalgic, and internet-driven. It reflected a generation shaping identity in a fast-moving, hyperconnected world.
Gen Z also known as Zoomers was the first generation to grow up with technolgy as an established commodity. Gen Z’s approach to love and dating is shaped by digital culture. The most popular song for the Gen Z was Iris by The Goo Goo Dolls which was made in 1998. This song reflects extreme vulnerability, longing, and willingness to sacrifice to be loved.
top10_songs_1997 <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 1997 & album_release_year <= 2012,
!is.na(track_popularity)) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
slice_max(track_popularity, n = 10, with_ties = FALSE) |>
ggplot(aes(x = reorder(paste(track_name, "-", artist_name), track_popularity),
y = track_popularity,
fill = factor(album_release_year))) +
geom_col() +
coord_flip() +
scale_fill_brewer(palette = "RdPu") +
scale_y_continuous(limits = c(0, 100), breaks = seq(0, 100, by = 10)) +
labs(
title = "Top 10 most popular songs (1997-2012)",
x = "Track — Artist",
y = "Popularity score",
fill = "Year"
) +
theme_minimal() +
theme(legend.position = "right")
top10_songs_1997
This line chart shows the trend in the most popular song for every year between 1997-2012 (Gen Z).
# Filter the data
library(plotly)
genZ_data <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 1997 & album_release_year <= 2012,
!is.na(track_popularity))
genZ_top_song <- genZ_data |>
group_by(album_release_year) |>
slice_max(track_popularity, n = 1, with_ties = FALSE) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
ungroup() |>
arrange(album_release_year)
plot_ly(
data = genZ_top_song,
x = ~album_release_year,
y = ~track_popularity,
type = "scatter",
mode = "lines+markers",
text = ~paste(
"Year:", album_release_year,
"<br>Popularity:", track_popularity,
"<br>Top Song:", track_name,
"<br>Artist:", artist_name
),
hoverinfo = "text"
) |>
layout(
title = "Top song popularity by year (Gen Z: 1997-2012)",
xaxis = list(title = "Year", dtick = 1),
yaxis = list(title = "Track popularity", range = c(0, 100))
)Generation Alpha is the generation following Generation Z. Generation Alpha is more commonly linked with digital immersion, rapid content consumption, personalization, and growing up fully surrounded by social media and streaming platforms. One of the clearest ways this identity is taking shape is through music. Growing up, the sound that defines Generation Alpha’s formative years is highly polished pop, viral-driven hits, and emotionally expressive tracks from artists like Taylor Swift and Billie Eilish, alongside songs that spread quickly through TikTok and other short-form video platforms. This shows that their culture is shaped less by traditional radio or albums and more by algorithms, trends, and online communities. Unlike Gen Z, who experienced the rise of streaming, Gen Alpha has never known a world without instant access to music. Their songs are designed to be shareable, visually connected, and emotionally direct. It reflects a generation forming identity in a fully digital, constantly connected world where music spreads at the speed of the internet.
Gen Alpha were born at time of failing fertility rates, and experienced the effect of COVID-19 as young children. Changes in the use of technology in classrooms and other aspects have had effect on how this generation experienced learning. The most popular song for Gen Alpha was Golden by HUNTR/X which was made in 2025. This song is about personal growth, self-discovery, and embracing ones true identity.
top10_songs_2013 <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 2013 & album_release_year <= 2025,
!is.na(track_popularity)) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
slice_max(track_popularity, n = 10, with_ties = FALSE) |>
ggplot(aes(x = reorder(paste(track_name, "-", artist_name), track_popularity),
y = track_popularity,
fill = factor(album_release_year))) +
geom_col() +
coord_flip() +
scale_fill_brewer(palette = "RdPu") +
scale_y_continuous(limits = c(0, 100), breaks = seq(0, 100, by = 10)) +
labs(
title = "Top 10 most popular songs (2013-2025)",
x = "Track — Artist",
y = "Popularity score",
fill = "Year"
) +
theme_minimal() +
theme(legend.position = "right")
top10_songs_2013
This line chart shows the trend in the most popular song for every year between 2013-2025 (Gen Alpha).
library(plotly)
genAlpha_data <- spotify |>
mutate(
album_release_year = as.integer(substr(album_release_date, 1, 4))
) |>
filter(album_release_year >= 2013 & album_release_year <= 2025,
!is.na(track_popularity))
genAlpha_top_song <- genAlpha_data |>
group_by(album_release_year) |>
slice_max(track_popularity, n = 1, with_ties = FALSE) |>
select(album_release_year, track_name, artist_name, track_popularity) |>
ungroup() |>
arrange(album_release_year)
plot_ly(
data = genAlpha_top_song,
x = ~album_release_year,
y = ~track_popularity,
type = "scatter",
mode = "lines+markers",
text = ~paste(
"Year:", album_release_year,
"<br>Popularity:", track_popularity,
"<br>Top Song:", track_name,
"<br>Artist:", artist_name
),
hoverinfo = "text"
) |>
layout(
title = "Top song popularity by year (Gen Alpha: 2013-2025)",
xaxis = list(title = "Year", dtick = 1),
yaxis = list(title = "Track popularity", range = c(0, 100))
)Out of the numerous genres in the data set the top 5 genres are soundtrack, country, pop, indie, and folk, soft pop, and rap. Soundtrack encompasses a wide variety of musical styles and genres and is an essential component of film and telvision production. Country, pop, indie, and folk is also a popular genre and has a overlapping themes which is why some artists categorized their songs having all 4 of these genres. Soft pop is a melodic soothing subgenre of pop that emphasizes gentle vocals and relaxed sound. Rap is a musical style characterized by rhythmic and rhyming speech and is a central element of hip-hop culture. Soundtrack is the most frequent genre with pop and the genre country, pop, indie and folk coming closely behind.
genre_top5 <- spotify |>
filter(!is.na(artist_genres), artist_genres != "N/A") |>
group_by(artist_genres) |>
summarise(n = n()) |>
slice_max(n, n = 5)
genre_top5 |>
ggplot(aes(x = reorder(artist_genres, n), y = n, fill = artist_genres)) +
geom_col() +
coord_flip() +
labs(
title = "Top 5 genres by frequency",
x = "Genre",
y = "Count",
fill = "Genre"
) +
theme_minimal() +
theme(legend.position = "none")
Out of the the three top genres Soundtrack is the only genre that has data from 1965-2025. For country, folk, indie, and pop the data doesn’t begin until 2005 and the genre pop doesn’t have data until 2000. Pop has the most linear growth increasing in popularity as the years go by. Soundtrack has a slight dip of popularity around 2015 and country, indie, folk, and pop has a slight boost of popularity around 2017 and 2025.
library(patchwork)
cpif_scatterplot<-
spotify|>
filter(!is.na(artist_genres), artist_genres != "N/A",
!is.na(track_popularity), track_popularity != "N/A") |>
mutate(
album_release_year = as.integer(
substr(album_release_date, 1, 4)))|>
filter(artist_genres == "country, pop, indie, folk")|>
ggplot(aes(x=album_release_year, y=track_popularity))+
geom_point(alpha=0.3)+
geom_smooth()+
labs(
title = "Track popularity of Country, Pop, Indie, and Folk",
x = "Album Release Year",
y = "Track popularity"
)+
scale_y_continuous(
limits = c(0, 100),
breaks = seq(0, 100, by = 25)
) +
scale_x_continuous(
limits = c(1965, 2025),
breaks = seq(1965, 2025, by = 10)
) +
annotate(
geom = "label", x = 1965, y = 95,
label = "Data doesn't start until around 2005 (GEN Z)",
hjust = "left", color = "red"
) +
annotate(
geom = "segment",
x = 1989, y = 70, xend = 2004, yend = 60,
color = "red",
arrow = arrow(type = "closed")
) +
theme_classic()
soundtrack_scatterplot<-
spotify|>
filter(!is.na(artist_genres), artist_genres != "N/A",
!is.na(track_popularity), track_popularity != "N/A") |>
mutate(
album_release_year = as.integer(
substr(album_release_date, 1, 4)))|>
filter(artist_genres == "soundtrack")|>
ggplot(aes(x=album_release_year, y=track_popularity))+
geom_point(alpha=0.3)+
geom_smooth()+
labs(
title = "Track popularity of Soundtrack",
x = "Album Release Year",
y = "Track popularity"
)+
scale_y_continuous(
limits = c(0, 100),
breaks = seq(0, 100, by = 25)
) +
scale_x_continuous(
limits = c(1965, 2025),
breaks = seq(1965, 2025, by = 10)
) +
theme_classic()
pop_scatterplot<-
spotify|>
filter(!is.na(artist_genres), artist_genres != "N/A",
!is.na(track_popularity), track_popularity != "N/A") |>
mutate(
album_release_year = as.integer(
substr(album_release_date, 1, 4)))|>
filter(artist_genres == "pop")|>
ggplot(aes(x=album_release_year, y=track_popularity))+
geom_point(alpha=0.3)+
geom_smooth()+
labs(
title = "Track popularity of Pop",
x = "Album Release Year",
y = "Track popularity"
)+
scale_y_continuous(
limits = c(0, 100),
breaks = seq(0, 100, by = 25)
) +
scale_x_continuous(
limits = c(1965, 2025),
breaks = seq(1965, 2025, by = 10)
) +
annotate(
geom = "label", x = 1965, y = 95,
label = "Data doesn't start until around 2000 (GEN Z)",
hjust = "left", color = "red"
) +
annotate(
geom = "segment",
x = 1985, y = 70, xend = 1998, yend = 60,
color = "red",
arrow = arrow(type = "closed")
) +
theme_classic()
soundtrack_scatterplot/cpif_scatterplot/pop_scatterplot
The top genre by frequency is Soundtrack. Soundtrack features a wide range of musical styles and is classified as a very diverse genre. The peak frequency for this genre is a track popularity of 40-49 with a frequency of 104.
soundtrack_popularity<-
spotify|>
filter(!is.na(track_popularity), track_popularity != "N/A",!is.na(artist_genres), artist_genres != "N/A")|>
filter(artist_genres == "soundtrack")
library(plotly)
plot_ly(
data = soundtrack_popularity,
x = ~track_popularity,
type = "histogram",
nbinsx = 10
)%>%
layout(
xaxis = list(title = "Track Popularity"),
yaxis = list(title = "Frequency")
)Taylor Swift has the most released tracks having released 7,505 total tracks from 2009-2025. The Weeknd comes in second place with 2649 total tracks released. With 12 total tracks Elle King has least number of tracks released.
spotify|>
mutate(
album_release_year = as.integer(
substr(album_release_date, 1, 4)))|>
filter(album_release_year >= 2009 & album_release_year <= 2025)|>
select(artist_name, album_total_tracks)|>
group_by(artist_name) |>
summarise(total_tracks = sum(album_total_tracks, na.rm = TRUE)) |>
arrange(desc(total_tracks))# A tibble: 2,136 × 2
artist_name total_tracks
<chr> <dbl>
1 Taylor Swift 7505
2 The Weeknd 2649
3 Lana Del Rey 1602
4 Drake 1578
5 Post Malone 1359
6 Ariana Grande 1287
7 Lady Gaga 1220
8 Bad Bunny 1156
9 The Neighbourhood 1060
10 Playboi Carti 982
# ℹ 2,126 more rowsTaylor Swift has released the most tracks from 20009-2025. Taylor Swift has significantly more tracks released than any other artist in this data-set.
spotify |>
mutate(
album_release_year = as.integer(
substr(album_release_date, 1, 4))) |>
filter(album_release_year >= 2009 & album_release_year <= 2025) |>
select(artist_name, album_total_tracks) |>
group_by(artist_name) |>
summarise(total_tracks = sum(album_total_tracks, na.rm = TRUE)) |>
arrange(desc(total_tracks)) |>
slice_max(total_tracks, n = 10) |>
ggplot(aes(x = reorder(artist_name, total_tracks), y = total_tracks, fill = artist_name)) +
geom_col() +
coord_flip() +
labs(
title = "Top 10 artists by total tracks (2009-2025)",
x = "Artist",
y = "Total tracks",
fill = "Artist"
) +
theme_minimal() +
theme(legend.position = "none")
Since Taylor Swift has released the most total tracks in this data-set, we chose to examine the popualrity of Taylor Swift’s explicit vs non explicit racks. It shows that Taylor Swifts explicit tracks had a higher median and more outliers on the higher end compared to the non explicit tracks. Her non explicit tracks also have a higher IQR(interquartile range). Her explicit tracks have a mean of 72.3 and a median of 69.5. Her non explicit tracks have a mean of 58 and a median of 62.
ts_explicit<- spotify|>
filter(!is.na(explicit), explicit != "N/A",
!is.na(artist_name), artist_name != "N/A",
!is.na(track_popularity), track_popularity != "N/A") |>
filter(artist_name== "Taylor Swift")|>
ggplot(aes(x =track_popularity , y = artist_name)) +
geom_boxplot(alpha = 0.05)+
facet_wrap(~ explicit)+
labs(title= "Track Popularity Taylor Swift's Explicit vs Non Explicit Songs",
x=' Track Popularity',
y='Artist')+
theme_bw()
ts_explicit
Music has changed a lot as different generations pass. Most of the popular songs of each generation reflects the time period and political and social climate.
The genre soundtrack is the most popular genre with data from 1965 to 2025.
The genre country, indie, folk, and pop overall has grown with slight boost in popularity in 2017 and 2025.
The genre pop has grown in popularity starting in the year 2000.
Taylor Swift has been dominating music by releasing the most tracks by far. Her explicit tracks have more popularity than her non explicit tracks.
Thanks for visiting our page!
msierra6@students.kennesaw.edu
mwalk249@students.kennesaw.edu