Before I start with my project, I would like to mention two things: First, that the lyrics and political candidates within the project, came from a github data set, link here: https://github.com/fivethirtyeight/data/blob/master/hip-hop-candidate-lyrics/genius_hip_hop_lyrics.csv Also, for fun, I would like to encourage you to look over some of the lyrics and see how the rappers incorporate each candidate. If you have time to spare!

Secondly, this idea was inspired by a creator on github and her article, link here: https://projects.fivethirtyeight.com/clinton-trump-hip-hop-lyrics/

This is my second document project on RMarkdown. I decided to look at the view of political candidates rapped in different songs within the Hip Hop music genre. I also wanted to take a look at how a couple of the candidates have changed within rap lyrics over the years. For example, have they increased, what years have spiked and why, and who’s mainly rapping who. The rap songs I will be look at range between the years of 1989 and 2016. The specific candidates that my data set includes, are: Donald Trump, Hillary Clinton, Jeb Bush, Bernie sanders, Chris Christie, Mike Huckabee, Ben Carson, and Ted Cruz.

Since we’re only looking at these candidates, my number one, go-to hypothesis would be that Donald Trump will be talked about a lot more than the other candidates. With that being said though, I am interested to see how exactly these rappers talk about him in their songs. Is it in a negative, positive, or neutral way; what words would be associated with him, especially if he’s being used in a positive way?

I essentially would like to know though, despite the rap community’s beliefs and values on him as a political candidate. If we solely look at the music and message of rap, has Hip Hop turned on Donald Trump or has he increased, decreased, or remained the same?

Let’s look at some visualizations and see!

library(tidyverse)
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
## ✓ ggplot2 3.3.5     ✓ purrr   0.3.4
## ✓ tibble  3.1.6     ✓ dplyr   1.0.7
## ✓ tidyr   1.2.0     ✓ stringr 1.4.0
## ✓ readr   2.1.2     ✓ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()
library(tidytext)
library(textdata)
library(wordcloud2)
library(readxl)
hiphoplyrics <- read_excel("/Applications/applied media analytics/hiphoplyrics.xlsx")

The first illustration I decided to look at was a zoomed out visualization of all the candidates and the number of times they have been used between the designated years (1989-2016). From the start, we can see that my hypothesis is correct, and more than likely yours too, in the sense that he is mentioned way more within the rap community than the other candidates.

hiphoplyrics %>% 
  count(candidate, sort = TRUE)
## # A tibble: 8 × 2
##   candidate           n
##   <chr>           <int>
## 1 Donald Trump      268
## 2 Hillary Clinton    92
## 3 Jeb Bush            9
## 4 Bernie Sanders      2
## 5 Chris Christie      2
## 6 Mike Huckabee       2
## 7 Ben Carson          1
## 8 Ted Cruz            1

As you can see, Donald Trump sweeps, with Hillary Clinton being second. Although she doesn’t have that many mentions compared to Donald Trump, she is used a significantly greater amount of times than the of the candidates.

With that being said, I will now shift my focus to only Trump and Clinton. I still would like to answer my previous stated questions about, “who is using them,” “when and where are the shifts in time,” “are they being used positively, negatively, or neutral and how much of each.” Adding in Clinton will be better for me as well because it will allow me to expand my investigation further. I immediately started thinking of more questions like, “is there a difference in how the rap community is talking about these two?” “is there a difference in rapping when it comes to a male and a female?” For example, “what’s negatively/positively being said about a man compared to a woman?”

Let’s look at some more visualizations and find out!

Before anything, I would just like to show you a timeline of each of their mentions and give insight for a couple of their peaks.

hiphoplyrics %>% 
  count(candidate, album_release_date, sort = TRUE) %>% 
  filter(candidate %in% c('Donald Trump','Hillary Clinton')) %>% 
  ggplot( aes(album_release_date, n, color = candidate)) + geom_line() + ylab("Number of Times Mentioned")

Now, side by side, I will do my best to try and analyze each of the peaks in the lines for each candidate. Essentially, I will try to explain why they started being talked about more in the years they had increases.

Let’s take a look at Hillary Clinton first. I would like to note that I researched all these facts, and I’m not saying this is why rappers rapped about her more because of the news I am about to list. Rather, I’m trying to come to some sense/reasoning as to why they did.

In order, everything pertaining to Hillary that I believe is in correlation with the times mentioned in rap songs, goes as followed: 1. In 1993, Hillary gets first introduced to the world though her husband Bill Clinton, who was elected president in this year. 2. In 1995, I believe she stayed relevant in the rap world because between 1995-1997 Bill Clinton was exposed for having an affair with, White House intern at the time Monica Lewinsky. 3. Around 2000, we can see she has a slight increase; I believe this is because this was the same year she was elected as the first female senate. Along with this, the following year, this was the last year Bill was in office. 4. Finally, the last peaks happened between 2009 and 2013. In 2009, she was elected promoted to secretary of state all the way through 2013. Therefore, she probably did news worthy actions that took place.

Now, for Donald Trump he’s in a different boat than Hillary. His family is already well known because of their wealth, so he’s already been in the public eye for a long time. Again, I’m not exactly sure why he was rapped about more in the years that spiked, but I researched and tried to find the most news worthy content that came up pertaining to Trump. With this being said though, Trump has done many things throughout the years that could lead to this spike, but for the year specifically, I will try to make some sense for it.

In order, everything concering Trump that I believe is in correlation with the times mentioned in rap songs, goes as followed: 1. In 1989, Trump had a very big controversy with the Central Park 5 controversy. IF you are not aware of this incident, in this year there were 5 black men that were convicted of rape. At the time, Trump said harsh comments towards them; it turned out that the men were wrongfully convicted. Since this day he’s never apologized for the comments, or felt that he was wrong. 2. In 1991, Trump claimed to have stopped a baseball bat-wielding attacker in New York. At first I thought that this was bizarre, but looking over the lyrics there are mentions of a bat.Another big thing during this year was that Trump was guilty of conspiring to avoid paying the union. 3. In 1998, Trump made headlines by saying that when he runs for office, he would run as a Republican because republicans are “dumb.” 4. In 2002, Trump said derogatory and controversial comments towards Iraq and their people. 5. In 2006, Trump furthered his wealth by turning into real estate mogul, had a new wife, a new baby boy, and had a reality TV show all in one year. 6. In 2011, Trump confused and hurt many people by not attending White House dinner. In the same year he said that President Obama would start a war against Iran. Again, I felt that this was bizarre, but there were mentions of Obama and Iran within the rappers lyrics. 7. In 2012, Trump was speculated to run. So, I think that based on his history, everyone was stunned. 8. In 2013, there was speculation that Trump spent the night in Russia because at this time he was getting close to Russia and Putin. 9. Finally, in 2015 (and the data only goes to 2016), Trump announced his White House run.

When comparing the two, I also find it fascinating how they both had speculations to run which made them popular in the news. But for Trump, rappers still focused on his money, his luxury, and wealth. Whereas for Hillary, they attacked her relationship with her husband, and their cheating scandal, even though she really didn’t do anything to hurt anyone as much as Trump.

Overall, I thought it would be interesting to breakdown some sort of events in order for both candidates to see a correlation of rappers and their lyrics and what happened in their lifetimes.

For the next visualizations I have put in two simple graphs, one for Trump and Hillary, that shows the top 10 artists tha are rapping about each them and the amount of times they have mentioned them in one of their songs.

hiphoplyrics %>% 
  filter(candidate %in% 'Donald Trump') %>% 
  count(artist, sort = TRUE)
## # A tibble: 192 × 2
##    artist         n
##    <chr>      <int>
##  1 Rick Ross      9
##  2 Nas            8
##  3 Migos          6
##  4 Young Thug     6
##  5 Lil Wayne      5
##  6 Raekwon        5
##  7 Shyne          4
##  8 Jay Z          3
##  9 Jeezy          3
## 10 Kanye West     3
## # … with 182 more rows

Since this is broken down, just by looking at the numbers we can see that these rappers don’t have an outstanding amount of times that mention trump, but we clearly Rick Ross has the highest amount.

hiphoplyrics %>% 
  filter(candidate %in% 'Hillary Clinton') %>% 
  count(artist, sort = TRUE)
## # A tibble: 79 × 2
##    artist          n
##    <chr>       <int>
##  1 Ice Cube        3
##  2 Angel Haze      2
##  3 Big Boi         2
##  4 Chino XL        2
##  5 Eminem          2
##  6 Gorilla Zoe     2
##  7 Gucci Mane      2
##  8 Jay Z           2
##  9 Ludacris        2
## 10 Rhymefest       2
## # … with 69 more rows

We already know that Hillary Clinton has a way less amount of times mentioned by rappers, but by breaking it down here there’s really a low number, with everyone basically being at the same number. With that being said, we clearly see that Ice Cube has the highest amount.

After looking at these two, I thought it was interesting how Trump is being mentioned by more new school rappers, where as Hillary is being mentioned by old school rappers. Not only does it show that Trump is still being mentioned throughout the decades, but it indirectly shows that what him as a candidate produces to the world is still being talked about.

I felt like these two visualizations were a big impact in the sense that Trump has about 250 mentions, and Hillary with 98, but by really dividing it up this way is helpful to give us a clear distinction.

My next visualizations will be two separate line graphs for Trump and Hillary; showing throughout the years how much each candidate is being talked about in a positive, negative, and neutral way. As well as when did each of those sentiments increase between the years.

hiphoplyrics %>% 
  filter(candidate %in% 'Donald Trump') %>% 
  group_by(album_release_date, sentiment) %>% 
  summarize(n = n()) %>% 
  ggplot(aes(album_release_date, n,color = sentiment)) + geom_line()
## `summarise()` has grouped output by 'album_release_date'. You can override using
## the `.groups` argument.

hiphoplyrics %>% 
  filter(candidate %in% 'Hillary Clinton') %>% 
  group_by(album_release_date, sentiment) %>% 
  summarize(n = n()) %>% 
  ggplot(aes(album_release_date, n,color = sentiment)) + geom_line()
## `summarise()` has grouped output by 'album_release_date'. You can override using
## the `.groups` argument.

What I find interesting about these two when comparing, is that Trump has more of a positive sentiment, where as Hillary has more of a negative sentiment. Why is this? Especially, when earlier in my analysis the more words associated with Hillary in a negative sense, pertained to her husband and the affair he had with Monica Lewinsky. In contrast with Trump, the words associated with negativity actually attacked his character.

Now, I will be showing you all the words associated with Trump in a negative way, with the bigger and bolder words being the most use of each word. Along with this I have included a table that gives you the exact number for each word.

First for Donald Trump:

hiphoplyrics %>% 
  filter(sentiment %in% 'negative') %>% 
  filter(candidate %in% 'Donald Trump') %>% 
  unnest_tokens(word, line) %>% 
  anti_join(stop_words) %>% 
  count(word, sort = TRUE) %>% 
  filter(!word %in% c('donald', 'trump')) %>% 
  wordcloud2()
## Joining, by = "word"
hiphoplyrics %>% 
  unnest_tokens(word,line) %>% 
  anti_join(stop_words) %>% 
  filter(sentiment %in% 'negative') %>% 
  filter(candidate %in% 'Donald Trump') %>% 
  count(word, sort = TRUE) %>%
  filter(!word %in% c("donald", "trump")) %>% 
  filter(n>2) %>% 
  knitr::kable() 
## Joining, by = "word"
word n
fuck 8
hair 3

Like mentioned before, we see that there are some racial undertones going on within this visualization. I would also like to note the size of this cloud, I would have suspected the visual filled up a little bit more.

Next is the same concept, but with the positive words.

hiphoplyrics %>% 
  filter(sentiment %in% 'positive') %>% 
  filter(candidate %in% 'Donald Trump') %>% 
  unnest_tokens(word, line) %>% 
  anti_join(stop_words) %>% 
  count(word, sort = TRUE) %>% 
  filter(!word %in% c('donald', 'trump')) %>% 
  wordcloud2()
## Joining, by = "word"
hiphoplyrics %>% 
  unnest_tokens(word,line) %>% 
  anti_join(stop_words) %>% 
  filter(sentiment %in% 'positive') %>% 
  filter(candidate %in% 'Donald Trump') %>% 
  count(word, sort = TRUE) %>%
  filter(!word %in% c("donald", "trump")) %>% 
  filter(n>2) %>% 
  knitr::kable() 
## Joining, by = "word"
word n
tower 19
money 16
towers 12
call 10
top 7
black 6
shit 6
floor 5
hood 5
plaza 5
rich 5
suite 5
cash 4
couple 4
livin 4
paid 4
pockets 4
real 4
uh 4
1 3
bill 3
bitches 3
dollar 3
don 3
estate 3
feel 3
floors 3
gates 3
international 3
million 3
night 3
penthouse 3
richer 3
talk 3
white 3

Firstly, within the cloud and table you can see what rappers consider “important.” Whether or not these are the only things that rappers could come up with in a positive way towards Trump, but here we can see that wealth and riches is a main feature within their raps. Secondly, I would like to note how there are way more words associated to Trump in a positive way compared to the negatives.

Not exactly sure why this would be, considering how intense the negative words came out to be, but maybe there are more ways you can explain someone’s wealth? Or maybe rappers are associating Trump just to what he’s known for, for example, Trump Towers is such a notorious name in the sense of money and wealthiness. Therefore maybe it’s not really him they’re referencing but his family instead.

Although the knit table for the negative words don’t really reflect the word cloud, it still emphasizes the point that there are more positives than negatives when talking about Trump.

Now onto to Hillary!

hiphoplyrics %>% 
  filter(sentiment %in% 'negative') %>% 
  filter(candidate %in% 'Hillary Clinton') %>% 
  unnest_tokens(word, line) %>% 
  anti_join(stop_words) %>% 
  count(word, sort = TRUE) %>% 
  filter(!word %in% c('hillary', 'clinton')) %>% 
  wordcloud2()
## Joining, by = "word"
hiphoplyrics %>% 
  unnest_tokens(word,line) %>% 
  anti_join(stop_words) %>% 
  filter(sentiment %in% 'negative') %>% 
  filter(candidate %in% 'Hillary Clinton') %>% 
  count(word, sort = TRUE) %>%
  filter(!word %in% c("hillary", "clinton", "hillary", "rodham")) %>% 
  filter(n>2) %>% 
  knitr::kable() 
## Joining, by = "word"
word n
bill 9
fuck 6
bitch 4
fucking 4
hilary 4
bottom 3
obama 3

For this visual, I found it interesting that there’s a good amount of negatives, but the most common words pertain to sexual slander and slander about Bill Clinton. So there’s nothing really challenging her character more so issues she’s going through.

hiphoplyrics %>% 
  filter(sentiment %in% 'positive') %>% 
  filter(candidate %in% 'Hillary Clinton') %>% 
  unnest_tokens(word, line) %>% 
  anti_join(stop_words) %>% 
  count(word, sort = TRUE) %>% 
  filter(!word %in% c('hillary', 'clinton')) %>% 
  wordcloud2()
## Joining, by = "word"
hiphoplyrics %>% 
  unnest_tokens(word,line) %>% 
  anti_join(stop_words) %>% 
  filter(sentiment %in% 'positive') %>% 
  filter(candidate %in% 'Hillary Clinton') %>% 
  count(word, sort = TRUE) %>%
  filter(!word %in% c("hillary", "clinton", "hillary", "rodham")) %>% 
  filter(n>2) %>% 
  knitr::kable() 
## Joining, by = "word"
word n
bills 4
feelin 3
stay 3

For this visual, I find it interesting that there are way less words connecting to Hillary, but a few of these words at least go with her character. With that being said, there are explicit wordings here, but overall I feel that it’s a solid positive list.

To compare knit tables, we can see better that there are way more negative words than positive for Hillary.

Overall I thought the comparisons between the two candidates were interesting. As we we can see, there is a difference when rapping about a female and a male negatively and positively. For Hillary, there’s nothing being negatively said about her, but more so her environment (her husband); the positives are going towards who she is but less of it. In contrast to Trump, there’s nothing being positively said about him, but his environment (family) and more of it. Whereas there are less negatives but going towards who he is and what he’s done.

To be a little bit more clear on the themes I’m picking up from the lyrics, I will give you an example of each sentiment for both candidates:

First for Hillary: Negatives: On January 1, 2013, rapper Bas released a song titled, “Shorty Baby.” The line that talks about Hillary is as followed: “Makin me call Lewinski, you fuckin up like Hillary did.” Just adding on to the fact, that nothing is being said negatively about her, but it’s a negative line that unfortunately has to do with her.

or

On January 1, 2008, Ludacris released a song, “Let’s Stay Together.” The line he had in association is, “Hillary’s still with Bill Clinton, how do they do it?” Although this isn’t too bad, it’s still a negative line towards Hillary because of Bill.

Positives: On January 1, 2007, rapper Trina released a song titled “Single Again (remix).” The line that she talks about Hillary is as followed: “Like Hillary, I’m the boss.” So, nothing about the affair, just all Hillary and who she is.

or

On January 1, 2010, rapper Big Sean released a song titled “Whatever You Want.” The line that he talks about Hillary is: “Leading lady like Hillary, grabbing big Bills.” Although yes, Bill is capitalized and probably has a deeper meaning, it’s nothing explicitly said about the affair and it’s more of hyping her up.

Although I was able to find a couple of examples for each, you can see there is a clear theme for each. The negatives (which there are more of) are about Bill and the positives are more of Women Empowerment (which there are less of).

Now for Trump: Positives: On January 1, 1995, artist 12 O’Clock - Brooklyn Zu made a song, “Protect Ya Neck II the Zoo,” and the line goes “Given the power punch, soon to be paid about Donald Trump.” Essentially saying he’s trying to get as rich as Trump because he’s very wealthy.

or

On January 1, 2002, 50 Cent made a song, “50/Banks;” his line goes “We can talk Trump talk, real estate, stocks, and bonds.” This line is really just saying that Trump is involved with so many things, which is why he as wealthy as he is.

Negatives: On January 1, 2005, artist Killah Priest, made a song “Right to Bare Arms.” The line where he mentions Trump goes, “This one’s for the cops and the war lord Trump.” So, this artist seemed to be going against what he’s done for the world

or

On January 1, 2005, artist High & Mighty released the song, “Dumb.” The line associated to Trump goes, “I’m Donald Trump, with raccoon-dos.” Not as bad as the first lyric, but again it’s strictly attacking his character.

Essentially for Trump, these lyrics are the opposite from Hillary’s lyrics. Again, there is a clear theme for each set of lyrics. The negatives (less of) are about his character and the positives are more of his luxurious life (which there are more of).

Coming into this project, I automatically thought that Trump was going to be mentioned more and have more negative sentiments. Ultimately my hypothesis was wrong, in both ways, but the research was very interesting. I got to see a history of the usage of Trump and Hillary through rap, who is mainly rapping these rappers, and how they’re rapping them. The clear takeaway from this whole study is the fact that there is a clear distinction on how you rap about men and women, no matter what their background is.

On that note, I would like to end you with an overall question of, why do you think here is such a difference when rapping about a female and a male in society? Along with that, what is the difference that’s causing this split?