eps plots

usc-sail · Sep 23, 2023 · ca58724 · ca58724
1 parent f4ab6c5
commit ca58724
Show file tree

Hide file tree

Showing 36 changed files with 491,688 additions and 12,471 deletions.
diff --git a/.RData b/.RData
diff --git a/.Rhistory b/.Rhistory
diff --git a/data/plots/age_trends.eps b/data/plots/age_trends.eps
diff --git a/data/plots/age_x_lang_trends.eps b/data/plots/age_x_lang_trends.eps
diff --git a/data/plots/age_x_skintone_trends.eps b/data/plots/age_x_skintone_trends.eps
diff --git a/data/plots/derogatory_trend.png b/data/plots/derogatory_trend.png
diff --git a/data/plots/derogatory_x_lang_trend.png b/data/plots/derogatory_x_lang_trend.png
diff --git a/data/plots/eps/age_trends.eps b/data/plots/eps/age_trends.eps
diff --git a/data/plots/eps/age_x_lang_trends.eps b/data/plots/eps/age_x_lang_trends.eps
diff --git a/data/plots/eps/age_x_skintone_trends.eps b/data/plots/eps/age_x_skintone_trends.eps
diff --git a/data/plots/eps/derogatory_trend.eps b/data/plots/eps/derogatory_trend.eps
diff --git a/data/plots/eps/derogatory_x_lang_trend.eps b/data/plots/eps/derogatory_x_lang_trend.eps
diff --git a/data/plots/eps/gender_trends.eps b/data/plots/eps/gender_trends.eps
diff --git a/data/plots/eps/gender_x_age_trends.eps b/data/plots/eps/gender_x_age_trends.eps
diff --git a/data/plots/eps/gender_x_age_x_skintone_trends.eps b/data/plots/eps/gender_x_age_x_skintone_trends.eps
diff --git a/data/plots/eps/gender_x_lang_trends.eps b/data/plots/eps/gender_x_lang_trends.eps
diff --git a/data/plots/eps/gender_x_skintone_trends.eps b/data/plots/eps/gender_x_skintone_trends.eps
diff --git a/data/plots/eps/gendername_trend.eps b/data/plots/eps/gendername_trend.eps
diff --git a/data/plots/eps/gendername_x_lang_trend.eps b/data/plots/eps/gendername_x_lang_trend.eps
diff --git a/data/plots/eps/religion_trend.eps b/data/plots/eps/religion_trend.eps
diff --git a/data/plots/eps/religion_x_lang_trend.eps b/data/plots/eps/religion_x_lang_trend.eps
diff --git a/data/plots/eps/skintone_trends.eps b/data/plots/eps/skintone_trends.eps
diff --git a/data/plots/eps/skintone_x_lang_trends.eps b/data/plots/eps/skintone_x_lang_trends.eps
diff --git a/data/plots/gender_trends.eps b/data/plots/gender_trends.eps
diff --git a/data/plots/gender_x_age_trends.eps b/data/plots/gender_x_age_trends.eps
diff --git a/data/plots/gender_x_age_x_skintone_trends.eps b/data/plots/gender_x_age_x_skintone_trends.eps
diff --git a/data/plots/gender_x_lang_trends.eps b/data/plots/gender_x_lang_trends.eps
diff --git a/data/plots/gender_x_skintone_trends.eps b/data/plots/gender_x_skintone_trends.eps
diff --git a/data/plots/gendername_trend.png b/data/plots/gendername_trend.png
diff --git a/data/plots/gendername_x_lang_trend.png b/data/plots/gendername_x_lang_trend.png
diff --git a/data/plots/religion_trend.png b/data/plots/religion_trend.png
diff --git a/data/plots/religion_x_lang_trend.png b/data/plots/religion_x_lang_trend.png
diff --git a/data/plots/skintone_trends.eps b/data/plots/skintone_trends.eps
diff --git a/data/plots/skintone_x_lang_trends.eps b/data/plots/skintone_x_lang_trends.eps
diff --git a/language.R b/language.R
@@ -1,5 +1,4 @@
-# set up ----------------------------------------------------------------------
-## load libraries ----
+# load libraries ----
 library(lme4)
 library(lmerTest)
 library(emmeans)
@@ -8,24 +7,23 @@ library(Kendall)
 
 emm_options(pbkrtest.limit = 28776, lmertest.limit = 28776)
 
-## read data ----
-## read long-form data (created and cleaned by @Sabyasachee)
+# read data ----
 lang_data <- read.csv("data/wide_lang.csv")
 lang_religion_data <- read.csv("data/long_lang_religion.csv")
 lang_gender_data <- read.csv("data/long_lang_gender.csv")
 
-# derogatory
+## derogatory ----
 lang_data$derogatory_per <- 100 * lang_data$derogatory / lang_data$transcript
 lang_data$derogatory_nonstop_per <-
   100 * lang_data$derogatory / lang_data$non_stopword_transcript
 
-# religion
+## religion ----
 lang_religion_data$name_count_per <- 
   100 * lang_religion_data$name_count / lang_religion_data$transcript
 lang_religion_data$name_count_nonstop_per <- 
   100 * lang_religion_data$name_count / lang_religion_data$non_stopword_transcript
 
-# gender
+## gender ----
 lang_gender_data$name_count_per <- 
   100 * lang_gender_data$name_count / lang_gender_data$transcript
 lang_gender_data$name_count_nonstop_per <- 
@@ -45,62 +43,68 @@ m_religion <- lmer(name_count_per ~ religion * year
                    + (1|program) + (1|lang) + (1|genre), lang_religion_data)
 m_religion_x_lang <- lmer(name_count_per ~ religion * lang * year 
                         + (1|program) + (1|genre), lang_religion_data)
+religion_trends <- emtrends(m_religion, ~religion, var = "year")
+religion_x_lang_trends <- emtrends(m_religion_x_lang, ~lang + religion, var = "year") %>%
+  as_tibble() %>% arrange(lang, religion)
 
 # gendered person names
 m_gendername <- lmer(name_count_per ~ gender * year 
                       + (1|program) + (1|lang) + (1|genre), lang_gender_data)
 m_gendername_x_lang <- lmer(name_count_per ~ gender * lang * year 
                            + (1|program) + (1|genre), lang_gender_data)
+gendername_trends <- emtrends(m_gendername, ~gender, var = "year")
+gendername_x_lang_trends <- emtrends(m_gendername_x_lang, ~lang + gender, var = "year") %>%
+  as_tibble() %>% arrange(lang, gender)
 
 ## plots ---
 # derogatory words
 ggplot(lang_data, aes(x = year, y = derogatory_per)) +
-  # geom_point(alpha = 0.5) +
+  geom_point() +
   geom_smooth(method = "lm", se = FALSE) +
   theme_minimal() +
   labs(y = "Percentage of Derogatory Words")
-ggsave("data/derogatory_trend.eps", device = "eps")
+ggsave("data/plots/derogatory_trend.png", device = "png")
 
 ggplot(lang_data, aes(x = year, y = derogatory_per, group = lang, 
                       color = lang)) +
-  # geom_point(alpha = 0.5) +
+  geom_point() +
   geom_smooth(method = "lm", se = FALSE) +
   theme_minimal() +
   labs(y = "Percentage of Derogatory Words")
-ggsave("data/derogatory_x_lang_trend.eps", device = "eps")
+ggsave("data/plots/derogatory_x_lang_trend.png", device = "png")
 
 # religion person names
 ggplot(lang_religion_data, aes(x = year, y = name_count_per, 
                                group = religion, color = religion)) +
-  # geom_point(alpha = 0.5) +
+  geom_point() +
   geom_smooth(method = "lm", se = FALSE) +
   theme_minimal() +
   labs(y = "Percentage of person names of different religions")
-ggsave("data/religion_trend.eps", device = "eps")
+ggsave("data/plots/religion_trend.png", device = "png")
 
 ggplot(lang_religion_data, aes(x = year, y = name_count_per, 
                                group = religion, color = religion)) +
-  # geom_point(alpha = 0.5) +
+  geom_point() +
   geom_smooth(method = "lm", se = FALSE) +
   theme_minimal() +
   labs(y = "Percentage of person names of different religions") +
   facet_wrap(~lang)
-ggsave("data/religion_x_lang_trend.eps", device = "eps")
+ggsave("data/plots/religion_x_lang_trend.png", device = "png")
 
 # gender person names
 ggplot(lang_gender_data, aes(x = year, y = name_count_per, 
                                group = gender, color = gender)) +
-  # geom_point(alpha = 0.5) +
+  geom_point() +
   geom_smooth(method = "lm", se = FALSE) +
   theme_minimal() +
   labs(y = "Percentage of person names of different genders")
-ggsave("data/gendername_trend.eps", device = "eps")
+ggsave("data/plots/gendername_trend.png", device = "png")
 
 ggplot(lang_gender_data, aes(x = year, y = name_count_per, 
                                group = gender, color = gender)) +
-  # geom_point(alpha = 0.5) +
+  geom_point() +
   geom_smooth(method = "lm", se = FALSE) +
   theme_minimal() +
   labs(y = "Percentage of person names of different genders") +
   facet_wrap(~lang)
-ggsave("data/gendername_x_lang_trend.eps", device = "eps")
+ggsave("data/plots/gendername_x_lang_trend.png", device = "png")