Taylor_Swift_RMD.Rmd

---
title: "Google_Capstone_Cyclistic"
author: "Aidan Rogers"
date: "`r Sys.Date()`"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

## R Markdown

### Divvy_Exercise_Full_Year_Analysis ###

# # # # # # # # # # # # # # # # # # # # # # # 
# Install required packages
# tidyverse for data import and wrangling
# lubridate for date functions
# ggplot for visualization
# # # # # # # # # # # # # # # # # # # # # # #  

library(tidyverse)  #helps wrangle data
library(lubridate)  #helps wrangle date attributes
library(ggplot2)  #helps visualize data
install.packages("ggrepel")
library(ggrepel)
getwd() #displays your working directory
setwd("/Users/aidan/Desktop/TaylorSwiftDA") #sets your working directory to simplify calls to data ... make sure to use your OWN username instead of mine ;)

theme_set(theme_bw())

#=====================
# STEP 1: COLLECT DATA
#=====================
# Upload Divvy datasets (csv files) here
taylor_data <- read_csv("spotify_taylorswift_excel_no_outlier.csv")

#====================================================
# STEP 2: WRANGLE DATA AND COMBINE INTO A SINGLE FILE
#====================================================
# Compare column names each of the files
# While the names don't have to be in the same order, they DO need to match perfectly before we can use a command to join them into one file
colnames(taylor_data)

# Rename columns  to make them consistent with q1_2020 (as this will be the supposed going-forward table design for Divvy)

(taylor_data <- rename(taylor_data
                   ,id = ID
                   ,track_number = `Track Number`
                   ,name = Name
                   ,album = Album
                   ,artist = Artist
                   ,release_date = `Release Date`
                   ,length = Length
                   ,popularity = Popularity
                   ,danceability = Danceability
                   ,acousticness = Acousticness
                   ,energy = Energy
                   ,instrumentalness = Instrumentalness
                   ,liveness = Liveness
                   ,loudness = Loudness
                   ,speechiness = Speechiness
                   ,valence = Valence
                   ,tempo = Tempo))

# Inspect the dataframes and look for incongruencies
str(taylor_data)


#======================================================
# STEP 3: CLEAN UP AND ADD DATA TO PREPARE FOR ANALYSIS
#======================================================
# Inspect the new table that has been created
colnames(taylor_data)  #List of column names
nrow(taylor_data)  #How many rows are in data frame?
dim(taylor_data)  #Dimensions of the data frame?
head(taylor_data)  #See the first 6 rows of data frame.  Also tail(all_trips)
str(taylor_data)  #See list of columns and data types (numeric, character, etc)
summary(taylor_data)  #Statistical summary of data. Mainly for numerics


table(taylor_data$album)


# Add columns that list the date, month, day, and year of each song
taylor_data$month <- format(as.Date(taylor_data$release_date), "%m")
taylor_data$day <- format(as.Date(taylor_data$release_date), "%d")
taylor_data$year <- format(as.Date(taylor_data$release_date), "%Y")
taylor_data$day_of_week <- format(as.Date(taylor_data$release_date), "%A")


# Inspect the structure of the columns
str(taylor_data)

# See how many songs Taylor released in each month and year
table(taylor_data$month)
table(taylor_data$year)



#=====================================
# STEP 4: CONDUCT DESCRIPTIVE ANALYSIS
#=====================================

# Put albums in order of release date
taylor_data$album <- ordered(taylor_data$album, levels=c("Taylor Swift", "Fearless (Platinum Edition)", "Speak Now (Deluxe Package)", "Red (Deluxe Edition)", "1989 (Deluxe)", "reputation", "Lover", "folklore (deluxe version)", "evermore (deluxe version)", "Fearless (Taylor's Version)", 
"Red (Taylor's Version)", "Midnights (3am Edition)"))

# Descriptive analysis on ride_length (all figures in seconds)
mean(taylor_data$popularity) #straight average (total ride length / rides)
median(taylor_data$popularity) #midpoint number in the ascending array of ride lengths
max(taylor_data$popularity) #longest ride
min(taylor_data$popularity) #shortest ride

# You can condense the four lines above to one line using summary() on the specific attribute
summary(taylor_data$popularity)
summary(taylor_data$loudness)


# Compare popularity for albums
aggregate(taylor_data$popularity ~ taylor_data$album, FUN = mean)
aggregate(taylor_data$popularity ~ taylor_data$album, FUN = median)
aggregate(taylor_data$popularity ~ taylor_data$album, FUN = max)
aggregate(taylor_data$popularity ~ taylor_data$album, FUN = min)

# Compare acousticness for albums
aggregate(taylor_data$acousticness ~ taylor_data$album, FUN = mean)
aggregate(taylor_data$acousticness ~ taylor_data$album, FUN = median)
aggregate(taylor_data$acousticness ~ taylor_data$album, FUN = max)
aggregate(taylor_data$acousticness ~ taylor_data$album, FUN = min)


# See what month each album was released and it's average level of popularity
aggregate(taylor_data$popularity ~ taylor_data$album + taylor_data$month, FUN = mean)

# Create a visualization for average popularity
taylor_data %>% 
  group_by(album) %>% 
  summarise(numer_of_songs = n(), average_popularity = mean(popularity)) %>% 
  arrange(album)  %>% 
  ggplot(aes(x = album, y = average_popularity, fill=album), title="Album Time") +
  geom_col(position = "dodge") + 
  theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
  labs(x="Album Name", y= "Average Popularity",
       title="Taylor Swift: Average Popularity of Albums") 
  
# Create a visualization for average tempo
taylor_data %>% 
  group_by(album) %>% 
  summarise(numer_of_songs = n(), average_tempo = mean(tempo)) %>% 
  arrange(album)  %>% 
  ggplot(aes(x = album, y = average_tempo, fill=album), title="Album Time") +
  geom_col(position = "dodge") + 
  theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
  labs(x="Album Name", y= " Average Tempo",
       title="Taylor Swift: Average Tempo of Albums")
       
       

# Create a Scatter Plot for Popularity vs Length
taylor_data %>%
  ggplot(aes(x = popularity, y = length)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Length",
       title="Popularity vs Length") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE)
  
# Create a Scatter Plot for Popularity vs Danceability
taylor_data %>%
  ggplot(aes(x = popularity, y = danceability)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Danceability",
       title="Popularity vs Danceability") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE)
  
# Create a Scatter Plot for Popularity vs Acousticness
taylor_data %>%
  ggplot(aes(x = popularity, y = acousticness)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Acousticness",
       title="Popularity vs Acousticness") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE)
  
# Create a Scatter Plot for Popularity vs Energy
taylor_data %>%
  ggplot(aes(x = popularity, y = energy)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Energy",
       title="Popularity vs Energy") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE)
  
# Create a Scatter Plot for Popularity vs Intrumentalness
taylor_data %>%
  ggplot(aes(x = popularity, y = instrumentalness)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Instrumentalness",
       title="Popularity vs Instrumentalness") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) 
  
# Create a Scatter Plot for Popularity vs Liveness
taylor_data %>%
  ggplot(aes(x = popularity, y = liveness)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Liveness",
       title="Popularity vs Liveness") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE)  

# Create a Scatter Plot for Popularity vs Loudness
taylor_data %>%
  ggplot(aes(x = popularity, y = loudness)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Loudness",
       title="Popularity vs Loudness") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) 
  
# Create a Scatter Plot for Popularity vs Speechiness
taylor_data %>%
  ggplot(aes(x = popularity, y = speechiness)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Speechiness",
       title="Popularity vs Speechiness") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) 
  
# Create a Scatter Plot for Popularity vs Valence
taylor_data %>%
  ggplot(aes(x = popularity, y = valence)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Valence",
       title="Popularity vs Valence") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) 
  
# Create a Scatter Plot for Popularity vs Tempo
taylor_data %>%
  ggplot(aes(x = popularity, y = tempo)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Tempo",
       title="Popularity vs Tempo") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE)
  
  
  
# Scatter Plot to show Release Date vs Popularity
taylor_data %>%
  ggplot(aes(x = release_date, y = popularity)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Release Date", y= "Popularity",
       title="Release Date vs Popularity") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) +
  scale_y_log10() 

# Same Scatter plot of Popularity vs Danceability but with some data labels
taylor_data %>%
  ggplot(aes(x = popularity, y = danceability)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Danceability",
       title="Popularity vs Danceability") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) +
  ggrepel::geom_text_repel(data = taylor_data %>%
                              filter(danceability< 0.32 | danceability > 0.83) %>%
                              sample(20),
                              aes(label=name))

# Same Scatter plot of Popularity vs Danceability but faceted to show each album
taylor_data %>%
  ggplot(aes(x = popularity, y = danceability)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Danceability",
       title="Popularity vs Danceability") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) +
  scale_x_log10() +
  facet_wrap(~album, ncol=3)  
  
# Create a Scatter Plot for energy vs acousticness
taylor_data %>%
  ggplot(aes(x = energy, y = acousticness)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Energy", y= "Acousticness",
       title="Energy vs Acousticness") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) +
  scale_x_log10() 
  
# Faceted Scatter plot to experiment with
taylor_data %>%
  ggplot(aes(x = popularity, y = danceability)) +
  geom_point(alpha=.8, aes(color=album)) +
  labs(x="Popularity", y= "Danceability",
       title="Popularity vs Danceability") +
  theme_bw(base_size = 10) +
  geom_smooth(method=lm,se=FALSE) +
  scale_x_log10() +
  facet_wrap(~album, ncol=3)