diff --git a/new_code.R b/new_code.R
new file mode 100644
index 0000000..98dd60c
--- /dev/null
+++ b/new_code.R
@@ -0,0 +1,182 @@
+
+
+pacman::p_load(tidyverse,
+               tidytext,
+               sentimentr,
+               topicmodels,
+               ggrepel,
+               kableExtra)
+
+source("code/01_calculate_sentiment.R")
+source("code/02_remove_stop_words.R")
+
+df_raw <- read_csv("data/raw/ham_lyrics.csv") %>% 
+  glimpse()
+
+set.seed(1234)
+
+
+# Data Sample -------------------------------------------------------------
+
+df_data_sample <- df_raw %>% 
+  sample_n(5) %>% 
+  mutate(
+    entry = row_number()
+  ) %>% 
+  select(entry, everything()) %>% 
+  rename(
+    Entry = entry,
+    Title = title,
+    Speaker = speaker,
+    Lines = lines
+  ) %>% 
+  mutate(
+    Speaker = str_to_title(Speaker),
+  )
+
+kable_data_sample <- df_data_sample %>% 
+  kable(., escape = TRUE, format = "latex") %>% 
+  row_spec(0, bold = TRUE) %>% 
+  column_spec(., column = 4, width = "6cm") %>%
+  save_kable(., file = "paper/tables/example_raw_data.tex")
+
+
+# Sentiment ---------------------------------------------------------------
+
+df_song_order <- df_raw %>% 
+  select(title) %>% 
+  distinct() %>% 
+  mutate(
+    song_number = row_number()
+  )
+
+df_sentimentr <- df_raw %>% 
+  mutate(
+    replace = gsub('[^ -~]', '', lines)
+  ) %>% 
+  pull(replace) %>% 
+  get_sentences() %>% 
+  sentiment() %>% 
+  as_tibble() %>% 
+  select(element_id, sentiment)
+
+df_sentimentr_sentiment <- df_raw %>% 
+  mutate(
+    element_id = row_number()
+  ) %>% 
+  left_join(., df_sentimentr, by = "element_id") %>% 
+  left_join(., df_song_order, by = "title") %>% 
+  group_by(song_number) %>% 
+  summarize(
+    sentimentr_sentiment = sum(sentiment)
+  )
+
+df_tokenized <- df_raw %>% 
+  left_join(., df_song_order, by = "title") %>% 
+  unnest_tokens(word, lines, to_lower = TRUE)
+
+
+afinn <- get_sentiments("afinn") %>% glimpse()
+bing <- get_sentiments("bing") %>% glimpse()
+nrc <- get_sentiments("nrc") %>% glimpse()
+loughran <- get_sentiments("loughran") %>% glimpse()
+
+df_afinn_sentiment <- df_tokenized %>% 
+  left_join(., afinn, by = "word") %>% 
+  group_by(song_number) %>% 
+  summarize(
+    afinn_sentiment = sum(value, na.rm = TRUE)
+  ) 
+
+df_bing_sentiment <- df_tokenized %>% 
+  left_join(., bing, by = "word") %>% 
+  mutate(
+    sentiment = case_when(
+      sentiment == "positive" ~ 1,
+      sentiment == "negative" ~ -1,
+      TRUE ~ NA_real_
+    )
+  ) %>% 
+  group_by(song_number) %>% 
+  summarize(
+    bing_sentiment = sum(sentiment, na.rm = TRUE)
+  ) 
+
+df_nrc_sentiment <- df_tokenized %>% 
+  left_join(., nrc, by = "word") %>% 
+  filter(sentiment %in% c("positive", "negative")) %>% 
+  mutate(
+    sentiment = case_when(
+      sentiment == "positive" ~ 1,
+      sentiment == "negative" ~ -1,
+      TRUE ~ NA_real_
+    )
+  ) %>% 
+  group_by(song_number) %>% 
+  summarize(
+    nrc_sentiment = sum(sentiment, na.rm = TRUE)
+  ) 
+
+df_sentiment <- list(df_afinn_sentiment, df_bing_sentiment, df_nrc_sentiment, df_sentimentr_sentiment) %>% 
+  reduce(left_join, by = "song_number") %>% 
+  pivot_longer(., cols = -song_number, names_to = "Sentiment Source", values_to = "Sentiment") %>% 
+  mutate(
+    "Sentiment Source" = str_remove(`Sentiment Source`, "_sentiment")
+  )
+
+df_sentiment %>% 
+  group_by(`Sentiment Source`) %>% 
+  mutate(
+    previous_song_sentiment = lag(Sentiment)
+  ) %>% 
+  ungroup() %>% 
+  mutate(
+    delta_sentiment = Sentiment - previous_song_sentiment
+  ) %>% 
+  select(song_number, delta_sentiment, `Sentiment Source`) %>% 
+  left_join(., df_song_order, by = "song_number") %>% 
+  group_by(`Sentiment Source`) %>% 
+  slice_max(delta_sentiment, n = 2)
+
+df_sentiment <- df_sentiment %>% 
+  left_join(., df_song_order, by = "song_number")
+
+p_sentiment <- ggplot(df_sentiment, aes(x = song_number, y = Sentiment)) +
+  geom_line() +
+  geom_hline(color = "red", yintercept = 0) +
+  geom_point() +
+  facet_wrap(~ `Sentiment Source`, ncol = 1, scales = "free_y") +
+  theme_minimal() +
+  labs(x = "Song Number") +
+  geom_text_repel(data = df_sentiment %>% filter(song_number %in% c(10, 11, 23)),
+             aes(label = title))
+  
+  
+ggsave(plot = p_sentiment, filename = "paper/figures/sentiment_by_stopwords.png")  
+
+
+
+Choose_Stopwords <- function(char_lexicon) {
+  
+  stop_words %>% 
+    filter(lexicon == char_lexicon)
+  
+}
+
+df_SMART_stop_words <- Choose_Stopwords("SMART") 
+df_onix_stop_words <- Choose_Stopwords("onix") 
+df_snowball_stop_words <- Choose_Stopwords("snowball") 
+
+Remove_Stopwords <- function(df_stop_words) {
+  
+  df_tokenized %>% 
+    anti_join(., df_stop_words, by = "word")
+  
+}
+
+df_SMART_filtered <- Remove_Stopwords(df_SMART_stop_words)
+df_onix_filtered <- Remove_Stopwords(df_onix_stop_words)
+df_snowball_filter <- Remove_Stopwords(df_snowball_stop_words)
+df_all_lexicons_filtered <- Remove_Stopwords(stop_words)
+
+write_csv(df_all_lexicons_filtered, "data/intermediate/tokenized_and_stop_words_removed.csv")