Soobin Choi 2023-05-01
knitr::opts_chunk$set(echo=TRUE, include=TRUE, comment="")
library(tidyverse)
library(dplyr)
library(ggplot2)original <- read_csv("data/original_num.csv")Rows: 458 Columns: 79
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): ResponseId, AT ever use, GENDER, GENDER_4_TEXT, SEXUALITY, AGE, RA...
dbl (71): LANGUAGE, BRO-F-parent, BRO-F-sibling, BRO-F-partner, BRO-F-cowork...
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
head(original)# A tibble: 6 × 79
ResponseId AT ev…¹ GENDER GENDE…² SEXUA…³ AGE LANGU…⁴ RACE OCCUP…⁵ BRO-F…⁶
<chr> <chr> <chr> <chr> <chr> <chr> <dbl> <chr> <chr> <dbl>
1 R_9FRUk95l… dude,m… Non-b… <NA> Gay 18-25 1 White Editor… NA
2 R_25EnKuXy… dude,g… Femin… <NA> Hetero 26-40 1 white Studen… NA
3 R_2pLWRKVY… bro,du… Femin… <NA> Bisexu… 26-40 1 White PhD st… 1
4 R_PBtJhKNv… bro,du… Mascu… <NA> Straig… 26-40 1 White Teacher 1
5 R_22wZhXw6… dude,m… Mascu… <NA> Straig… 41-55 1 White Civil … NA
6 R_305B45bv… bro,du… Femin… <NA> Bisexu… 18-25 1 White Econom… 1
# … with 69 more variables: `BRO-F-sibling` <dbl>, `BRO-F-partner` <dbl>,
# `BRO-F-coworker` <dbl>, `BRO-F-boss` <dbl>, `BRO-F-friend` <dbl>,
# `BRO-F-stranger` <dbl>, `BRO-M-parent` <dbl>, `BRO-M-sibling` <dbl>,
# `BRO-M-partner` <dbl>, `BRO-M-coworker` <dbl>, `BRO-M-boss` <dbl>,
# `BRO-M-friend` <dbl>, `BRO-M-stranger` <dbl>, `BRUH-F-parent` <dbl>,
# `BRUH-F-sibling` <dbl>, `BRUH-F-partner` <dbl>, `BRUH-F-coworker` <dbl>,
# `BRUH-F-boss` <dbl>, `BRUH-F-friend` <dbl>, `BRUH-F-stranger` <dbl>, …
# process the column for the analysis - term, addressee gender, relationship
# And, change the name of columns
original1 <- original %>%
pivot_longer(`BRO-F-parent`:`MAN-F-stranger`, names_to = "TERM", values_to = "FREQUENCY") %>%
separate(col = TERM, c("TERM", "ADDR-GENDER", "RANK"), sep = "-") %>%
rename(`RESP-ID` = ResponseId,
ATUSE = `AT ever use`,
`RESP-GENDER` = GENDER) %>%
mutate(RANK = str_to_upper(RANK))# check the categories in AGE column
original1 %>%
select(AGE) %>%
unique()# A tibble: 6 × 1
AGE
<chr>
1 18-25
2 26-40
3 41-55
4 56-70
5 Over 70
6 <NA>
original2 <- original1 %>%
# remove NAs in AGE column
filter(!is.na(AGE)) %>%
mutate(`RESP-GENDER` = str_to_title(`RESP-GENDER`)) %>%
relocate(`ADDR-GENDER`, .after="RESP-GENDER") %>%
# merge '56-70' and 'over 70' into one category
mutate_all(funs(str_replace_all(., "56-70", "Over 70"))) %>%
mutate_all(funs(str_replace_all(., "Over 70", "56-Over 70"))) %>%
# change 'other' gender to 'non-binary'
map_dfr(~ str_replace_all(., "Other", "Non-Binary"))Warning: `funs()` was deprecated in dplyr 0.8.0.
Please use a list of either functions or lambdas:
# Simple named list:
list(mean = mean, median = median)
# Auto named with `tibble::lst()`:
tibble::lst(mean, median)
# Using lambdas
list(~ mean(., trim = .2), ~ median(., na.rm = TRUE))
This warning is displayed once every 8 hours.
Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.
# make another column - clean up `SEXUALITY` column as binary value
org_clean <- original2 %>%
mutate(SEXUALITY2 = grepl("(het|straight)", tolower(original2$SEXUALITY))) %>%
relocate(SEXUALITY2, .after = SEXUALITY) %>%
map_dfr(~ str_replace_all(., c("TRUE" = "Hetero", "FALSE" = "Non-Hetero"))) %>%
mutate(FREQUENCY = as.numeric(FREQUENCY))Okay, I have left with 31,990 values in total.
# make new column called 'DUDE': TRUE means the participant uses *dude*, FALSE means they don't
dude_df <- org_clean %>%
mutate(DUDE = grepl("dude", org_clean$ATUSE))
head(dude_df, 10)# A tibble: 10 × 15
`RESP-ID` ATUSE RESP-…¹ ADDR-…² GENDE…³ SEXUA…⁴ SEXUA…⁵ AGE LANGU…⁶ RACE
<chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
2 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
3 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
4 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
5 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
6 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
7 R_9FRUk95l… dude… Non-Bi… F <NA> Gay Non-He… 18-25 1 White
8 R_9FRUk95l… dude… Non-Bi… M <NA> Gay Non-He… 18-25 1 White
9 R_9FRUk95l… dude… Non-Bi… M <NA> Gay Non-He… 18-25 1 White
10 R_9FRUk95l… dude… Non-Bi… M <NA> Gay Non-He… 18-25 1 White
# … with 5 more variables: OCCUPATION <chr>, TERM <chr>, RANK <chr>,
# FREQUENCY <dbl>, DUDE <lgl>, and abbreviated variable names ¹`RESP-GENDER`,
# ²`ADDR-GENDER`, ³GENDER_4_TEXT, ⁴SEXUALITY, ⁵SEXUALITY2, ⁶LANGUAGE
# Plot - Dude Usage by gender and age
dude_df %>%
group_by(`RESP-GENDER`, AGE, DUDE) %>%
summarise(Count = n_distinct(`RESP-ID`)) %>%
ggplot(aes(x = AGE, y = Count, fill = DUDE)) +
facet_wrap(vars(`RESP-GENDER`), ncol = 3) +
geom_col(position = "fill") +
labs(title = "Dude Usage by Respondants' Gender and Age")`summarise()` has grouped output by 'RESP-GENDER', 'AGE'. You can override
using the `.groups` argument.
'
dude_df %>%
filter(TERM == "DUDE" & `RESP-GENDER` != "Non-Binary") %>%
group_by(AGE, `ADDR-GENDER`, `RESP-GENDER`) %>%
summarise(count = n_distinct(`RESP-ID`)) %>%
unite(Gender, `RESP-GENDER`, `ADDR-GENDER`) %>%
ggplot(aes(x = AGE, weight = count, fill = Gender)) +
geom_bar(position = "dodge")
'[1] "\ndude_df %>% \n filter(TERM == \"DUDE\" & `RESP-GENDER` != \"Non-Binary\") %>% \n group_by(AGE, `ADDR-GENDER`, `RESP-GENDER`) %>% \n summarise(count = n_distinct(`RESP-ID`)) %>% \n unite(Gender, `RESP-GENDER`, `ADDR-GENDER`) %>% \n ggplot(aes(x = AGE, weight = count, fill = Gender)) +\n geom_bar(position = \"dodge\")\n"
# consider respondent's gender only - total count of people who answered that they use dude
dude_df %>%
group_by(`RESP-GENDER`, DUDE) %>%
summarise(Count = n_distinct(`RESP-ID`))`summarise()` has grouped output by 'RESP-GENDER'. You can override using the
`.groups` argument.
# A tibble: 6 × 3
# Groups: RESP-GENDER [3]
`RESP-GENDER` DUDE Count
<chr> <lgl> <int>
1 Feminine FALSE 74
2 Feminine TRUE 211
3 Masculine FALSE 28
4 Masculine TRUE 90
5 Non-Binary FALSE 10
6 Non-Binary TRUE 44
dude_df %>%
group_by(`RESP-GENDER`, DUDE) %>%
summarise(Count = n_distinct(`RESP-ID`)) %>%
ggplot(aes(x = `RESP-GENDER`, y = Count, fill = DUDE)) +
geom_col(position = "fill") +
ylab("Percentage") +
labs(title = "Dude Usage by Respondents' Gender")`summarise()` has grouped output by 'RESP-GENDER'. You can override using the
`.groups` argument.
dude_df %>%
filter(TERM == "DUDE") %>%
filter(!is.na(FREQUENCY)) %>%
group_by(`ADDR-GENDER`, `RESP-GENDER`) %>%
summarise(mean_freq = mean(FREQUENCY)) %>%
ggplot(aes(x = `RESP-GENDER`, y = mean_freq, col = `ADDR-GENDER`, group = `ADDR-GENDER`))+
geom_point() +
geom_text(aes(label = round(mean_freq, 2), vjust = -0.5, nudge_y = 0.1)) +
geom_path() +
labs(title = "Reported Frequency of Dude by Gender of Speaker and Addressee") +
ylim(1.9, 2.65)`summarise()` has grouped output by 'ADDR-GENDER'. You can override using the
`.groups` argument.
Warning: Ignoring unknown aesthetics: nudge_y
# sort people who use dude and/or bro
dudebro <- org_clean %>%
filter(grepl("dude", org_clean$ATUSE) == TRUE | grepl("bro", org_clean$ATUSE) == TRUE)# plot - dude/bro usage by rank and gender
dudebro %>%
filter(TERM %in% c("DUDE", "BRO")) %>%
group_by(`RESP-GENDER`, `ADDR-GENDER`, TERM, `RANK`) %>%
summarise(mean_freq = mean(as.numeric(FREQUENCY), na.rm = TRUE)) %>%
ggplot(aes(x = `RANK`, y = mean_freq, group = TERM, col = TERM)) +
facet_wrap(`ADDR-GENDER`~`RESP-GENDER`, ncol = 1, strip.position = "right",
labeller = labeller(
`ADDR-GENDER` = c(`F`="Addr_F", `M`="Addr_M"),
`RESP-GENDER` = c(`Feminine`="Resp_F", `Masculine`="Resp_M", `Non-Binary`="Non-Binary"))) +
geom_line() +
geom_point() +
scale_y_continuous(limits = c(0, 4)) +
labs(title = "Compare dude/bro based on interlocutors' gender", x = 'Relationship')`summarise()` has grouped output by 'RESP-GENDER', 'ADDR-GENDER', 'TERM'. You
can override using the `.groups` argument.
it contains too much information in one plot. need to make it more reader-friendly
# DUDE
dudebro %>%
filter(TERM == "DUDE") %>%
group_by(`RESP-GENDER`, `ADDR-GENDER`, TERM, `RANK`) %>%
summarise(mean_freq = mean(as.numeric(FREQUENCY), na.rm = TRUE)) %>%
unite(GENDER, `RESP-GENDER`, `ADDR-GENDER`)`summarise()` has grouped output by 'RESP-GENDER', 'ADDR-GENDER', 'TERM'. You
can override using the `.groups` argument.
# A tibble: 42 × 4
# Groups: TERM [1]
GENDER TERM RANK mean_freq
<chr> <chr> <chr> <dbl>
1 Feminine_F DUDE BOSS 1.32
2 Feminine_F DUDE COWORKER 2.31
3 Feminine_F DUDE FRIEND 3.04
4 Feminine_F DUDE PARENT 1.40
5 Feminine_F DUDE PARTNER 2.10
6 Feminine_F DUDE SIBLING 2.22
7 Feminine_F DUDE STRANGER 1.70
8 Feminine_M DUDE BOSS 1.34
9 Feminine_M DUDE COWORKER 2.44
10 Feminine_M DUDE FRIEND 3.19
# … with 32 more rows
dudebro %>%
filter(TERM == "DUDE") %>%
group_by(`RESP-GENDER`, `ADDR-GENDER`, TERM, `RANK`) %>%
summarise(mean_freq = mean(as.numeric(FREQUENCY), na.rm = TRUE)) %>%
unite(GENDER, `RESP-GENDER`, `ADDR-GENDER`) %>%
ggplot(aes(x = RANK, y = mean_freq, col = GENDER, group = GENDER)) +
geom_point() +
geom_line() +
xlab("Relationship") +
labs(title = "Dude Usage by Gender") +
ylim(0.5, 4)`summarise()` has grouped output by 'RESP-GENDER', 'ADDR-GENDER', 'TERM'. You
can override using the `.groups` argument.
# BRO
dudebro %>%
filter(TERM == "BRO") %>%
group_by(`RESP-GENDER`, `ADDR-GENDER`, TERM, `RANK`) %>%
summarise(mean_freq = mean(as.numeric(FREQUENCY), na.rm = TRUE)) %>%
unite(GENDER, `RESP-GENDER`, `ADDR-GENDER`)`summarise()` has grouped output by 'RESP-GENDER', 'ADDR-GENDER', 'TERM'. You
can override using the `.groups` argument.
# A tibble: 42 × 4
# Groups: TERM [1]
GENDER TERM RANK mean_freq
<chr> <chr> <chr> <dbl>
1 Feminine_F BRO BOSS 1.21
2 Feminine_F BRO COWORKER 1.86
3 Feminine_F BRO FRIEND 2.89
4 Feminine_F BRO PARENT 1.21
5 Feminine_F BRO PARTNER 1.71
6 Feminine_F BRO SIBLING 2.20
7 Feminine_F BRO STRANGER 1.57
8 Feminine_M BRO BOSS 1.19
9 Feminine_M BRO COWORKER 2.05
10 Feminine_M BRO FRIEND 3.19
# … with 32 more rows
dudebro %>%
filter(TERM == "BRO") %>%
group_by(`RESP-GENDER`, `ADDR-GENDER`, TERM, `RANK`) %>%
summarise(mean_freq = mean(as.numeric(FREQUENCY), na.rm = TRUE)) %>%
unite(GENDER, `RESP-GENDER`, `ADDR-GENDER`) %>%
ggplot(aes(x = RANK, y = mean_freq, col = GENDER, group = GENDER)) +
geom_point() +
geom_line() +
xlab("Relationship") +
labs(title = "Bro Usage by Gender") +
ylim(0.5, 4)`summarise()` has grouped output by 'RESP-GENDER', 'ADDR-GENDER', 'TERM'. You
can override using the `.groups` argument.
org_clean %>%
mutate(RACE = str_to_title(RACE)) %>%
group_by(RACE) %>%
summarize(n = n_distinct(`RESP-ID`))# A tibble: 82 × 2
RACE n
<chr> <int>
1 African American 1
2 Anglo 1
3 Anglo Australian (White) 1
4 Arab 1
5 Ashkenazi Jewish 1
6 Asian 7
7 Asian-American 1
8 Asian Indian 2
9 Asian, Korean 1
10 Australian 1
# … with 72 more rows
# load the data
rank_sent <- read_csv("data/rank_sent.csv")Rows: 458 Columns: 31
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (7): ResponseId, AT ever use, RESP-GENDER, GENDER-4-TEXT, SEXUALITY, AG...
dbl (24): MUSIC-BRO, MUSIC-GIRL, MUSIC-DUDE, MUSIC-BRUH, MEETING-BRO, MEETIN...
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
head(rank_sent)# A tibble: 6 × 31
Respon…¹ AT ev…² RESP-…³ GENDE…⁴ SEXUA…⁵ AGE MUSIC…⁶ MUSIC…⁷ MUSIC…⁸ MUSIC…⁹
<chr> <chr> <chr> <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
1 R_9FRUk… dude,m… Non-bi… <NA> Gay 18-25 2 4 1 3
2 R_25EnK… dude,g… Femini… <NA> Hetero 26-40 3 2 1 4
3 R_2pLWR… bro,du… Femini… <NA> Bisexu… 26-40 3 2 1 4
4 R_PBtJh… bro,du… Mascul… <NA> Straig… 26-40 2 3 1 4
5 R_22wZh… dude,m… Mascul… <NA> Straig… 41-55 2 3 1 4
6 R_305B4… bro,du… Femini… <NA> Bisexu… 18-25 3 2 1 4
# … with 21 more variables: `MEETING-BRO` <dbl>, `MEETING-GIRL` <dbl>,
# `MEETING-DUDE` <dbl>, `MEETING-BRUH` <dbl>, `SHUTUP-BRO` <dbl>,
# `SHUTUP-GIRL` <dbl>, `SHUTUP-DUDE` <dbl>, `SHUTUP-BRUH` <dbl>,
# `GOODTOSEE-BRO` <dbl>, `GOODTOSEE-GIRL` <dbl>, `GOODTOSEE-DUDE` <dbl>,
# `GOODTOSEE-BRUH` <dbl>, `SIGNAL-BRO` <dbl>, `SIGNAL-GIRL` <dbl>,
# `SIGNAL-DUDE` <dbl>, `SIGNAL-BRUH` <dbl>, `EMAIL-BRO` <dbl>,
# `EMAIL-GIRL` <dbl>, `EMAIL-DUDE` <dbl>, `EMAIL-BRUH` <dbl>, Q23 <chr>, …
rank_clean <- rank_sent %>%
select(-Q23) %>%
rename(`RESP-ID` = ResponseId,
ATUSE = `AT ever use`)colnames(rank_clean) [1] "RESP-ID" "ATUSE" "RESP-GENDER" "GENDER-4-TEXT"
[5] "SEXUALITY" "AGE" "MUSIC-BRO" "MUSIC-GIRL"
[9] "MUSIC-DUDE" "MUSIC-BRUH" "MEETING-BRO" "MEETING-GIRL"
[13] "MEETING-DUDE" "MEETING-BRUH" "SHUTUP-BRO" "SHUTUP-GIRL"
[17] "SHUTUP-DUDE" "SHUTUP-BRUH" "GOODTOSEE-BRO" "GOODTOSEE-GIRL"
[21] "GOODTOSEE-DUDE" "GOODTOSEE-BRUH" "SIGNAL-BRO" "SIGNAL-GIRL"
[25] "SIGNAL-DUDE" "SIGNAL-BRUH" "EMAIL-BRO" "EMAIL-GIRL"
[29] "EMAIL-DUDE" "EMAIL-BRUH"
rank_clean %>%
select(contains(c("ID", "MUSIC")))# A tibble: 458 × 5
`RESP-ID` `MUSIC-BRO` `MUSIC-GIRL` `MUSIC-DUDE` `MUSIC-BRUH`
<chr> <dbl> <dbl> <dbl> <dbl>
1 R_9FRUk95l1fWKDZf 2 4 1 3
2 R_25EnKuXyTReLPE2 3 2 1 4
3 R_2pLWRKVYVALdO6e 3 2 1 4
4 R_PBtJhKNv0JB10GJ 2 3 1 4
5 R_22wZhXw6DTa6krK 2 3 1 4
6 R_305B45bvuFeKZUx 3 2 1 4
7 R_vjZzVyscBlf29AB 3 2 1 4
8 R_1IL6Ojmmkco93Hi 2 4 1 3
9 R_0W1f5Y9lpxf2LWV 2 3 1 4
10 R_3CCJoj9CLivBnt1 2 4 1 3
# … with 448 more rows
meeting <- rank_clean %>%
select(contains(c("ID", "MEETING")))
meeting_final <- meeting %>%
rename_with(., ~gsub("MEETING-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
mutate(SENT = "Meeting", .after = TERM)
meeting %>%
rename_with(., ~gsub("MEETING-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
ggplot(aes(x = TERM, y = mean_rank)) +
geom_path(group = 1, color = "dark green") +
geom_point(color = "dark green") +
geom_text(aes(label = round(mean_rank, 2), hjust = -0.7, nudge_x = 0.1), color = "dark green") +
scale_y_reverse(limits = c(5, 0.5)) +
labs(title = "Mean Rank - \"When is the meeting?\"")Warning: Ignoring unknown aesthetics: nudge_x
music <- rank_clean %>%
select(contains(c("ID", "MUSIC")))
music_final <- music %>%
rename_with(., ~gsub("MUSIC-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
mutate(SENT = "Music", .after = TERM)shutup <- rank_clean %>%
select(contains(c("ID", "SHUTUP")))
shutup_final <- shutup %>%
rename_with(., ~gsub("SHUTUP-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
mutate(SENT = "Shut up", .after = TERM)goodtosee <- rank_clean %>%
select(contains(c("ID", "GOODTOSEE")))
goodtosee_final <- goodtosee %>%
rename_with(., ~gsub("GOODTOSEE-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
mutate(SENT = "Good to see", .after = TERM)signal <- rank_clean %>%
select(contains(c("ID", "SIGNAL")))
signal_final <- signal %>%
rename_with(., ~gsub("SIGNAL-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
mutate(SENT = "Signal", .after = TERM)email <- rank_clean %>%
select(contains(c("ID", "EMAIL")))
email_final <- email %>%
rename_with(., ~gsub("EMAIL-", "", .x)) %>%
pivot_longer(!`RESP-ID`, names_to = "TERM", values_to = "RANK") %>%
na.omit(RANK) %>%
group_by(TERM) %>%
summarise(mean_rank = mean(RANK)) %>%
mutate(SENT = "Email", .after = TERM)sent_final <- rbind(meeting_final, music_final, goodtosee_final, shutup_final, signal_final, email_final)
sent_final %>%
group_by(TERM) %>%
summarize(mean = mean(mean_rank))# A tibble: 4 × 2
TERM mean
<chr> <dbl>
1 BRO 2.33
2 BRUH 3.42
3 DUDE 1.39
4 GIRL 2.86
sent_final %>%
ggplot(aes(x = TERM, y = mean_rank, group = SENT, col = SENT)) +
geom_path() +
geom_point() +
scale_y_reverse(limits = c(4, 1))
scale_x_discrete(limits = c("DUDE", "BRO", "BRUH", "GIRL"))<ggproto object: Class ScaleDiscretePosition, ScaleDiscrete, Scale, gg>
aesthetics: x xmin xmax xend
axis_order: function
break_info: function
break_positions: function
breaks: waiver
call: call
clone: function
dimension: function
drop: TRUE
expand: waiver
get_breaks: function
get_breaks_minor: function
get_labels: function
get_limits: function
guide: waiver
is_discrete: function
is_empty: function
labels: waiver
limits: DUDE BRO BRUH GIRL
make_sec_title: function
make_title: function
map: function
map_df: function
n.breaks.cache: NULL
na.translate: TRUE
na.value: NA
name: waiver
palette: function
palette.cache: NULL
position: bottom
range: <ggproto object: Class RangeDiscrete, Range, gg>
range: NULL
reset: function
train: function
super: <ggproto object: Class RangeDiscrete, Range, gg>
range_c: <ggproto object: Class RangeContinuous, Range, gg>
range: NULL
reset: function
train: function
super: <ggproto object: Class RangeContinuous, Range, gg>
rescale: function
reset: function
scale_name: position_d
train: function
train_df: function
transform: function
transform_df: function
super: <ggproto object: Class ScaleDiscretePosition, ScaleDiscrete, Scale, gg>
library(tidytext)
library(stopwords)Warning: 패키지 'stopwords'는 R 버전 4.2.2에서 작성되었습니다
brobruh <- read_csv("data/brobruh.csv")Rows: 458 Columns: 7
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (7): ResponseId, AT ever use, RESP-GENDER, GENDER-4-TEXT, SEXUALITY, AGE...
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
head(brobruh)# A tibble: 6 × 7
ResponseId `AT ever use` `RESP-GENDER` GENDER…¹ SEXUA…² AGE Q23
<chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 R_9FRUk95l1fWKDZf dude,man Non-binary <NA> Gay 18-25 <NA>
2 R_25EnKuXyTReLPE2 dude,girl,man Feminine <NA> Hetero 26-40 Depe…
3 R_2pLWRKVYVALdO6e bro,dude,girl,man Feminine <NA> Bisexu… 26-40 Bruh…
4 R_PBtJhKNv0JB10GJ bro,dude,man Masculine <NA> Straig… 26-40 Bruh…
5 R_22wZhXw6DTa6krK dude,man Masculine <NA> Straig… 41-55 <NA>
6 R_305B45bvuFeKZUx bro,dude,girl,man Feminine <NA> Bisexu… 18-25 I do…
# … with abbreviated variable names ¹`GENDER-4-TEXT`, ²SEXUALITY
# tokenize the response by sentence, sort out the columns we don't need
brobruh2 <- brobruh %>%
rename(ANSWER = Q23) %>%
select(-c(`GENDER-4-TEXT`, `AT ever use`)) %>%
unnest_tokens(`SENT-TOK`, ANSWER, token = "sentences") %>%
# need to split the sents by comma, too.
unnest_tokens(`SENT-TOK2`, `SENT-TOK`, token = stringr::str_split, pattern = ",") %>%
na.omit(`SENT-TOK`)brobruh_clean <- brobruh2 %>%
mutate(BRO = grepl('bro', `SENT-TOK2`),
BRUH = grepl("bruh", `SENT-TOK2`)) %>%
relocate(BRO, BRUH, .before = `SENT-TOK2`)
bro_clean <- brobruh_clean %>%
filter(BRO == TRUE & BRUH == FALSE)
bruh_clean <- brobruh_clean %>%
filter(BRO == FALSE & BRUH == TRUE)
head(bro_clean)# A tibble: 6 × 7
ResponseId `RESP-GENDER` SEXUALITY AGE BRO BRUH `SENT-TOK2`
<chr> <chr> <chr> <chr> <lgl> <lgl> <chr>
1 R_305B45bvuFeKZUx Feminine Bisexual 18-25 TRUE FALSE " but do use…
2 R_vjZzVyscBlf29AB Masculine cis 41-55 TRUE FALSE "bro is more…
3 R_Rf6qeJ9SaOibzwJ Feminine Hetero 18-25 TRUE FALSE "bro is more…
4 R_2PofVys50lS49yH Masculine Straight 26-40 TRUE FALSE "bro is used…
5 R_2PofVys50lS49yH Masculine Straight 26-40 TRUE FALSE "\"bro"
6 R_3g4NXfQGEg4j3Vx Masculine heteroflexible 26-40 TRUE FALSE "\"bro\" is …
head(bruh_clean)# A tibble: 6 × 7
ResponseId `RESP-GENDER` SEXUALITY AGE BRO BRUH `SENT-TOK2`
<chr> <chr> <chr> <chr> <lgl> <lgl> <chr>
1 R_2pLWRKVYVALdO6e Feminine Bisexual queer 26-40 FALSE TRUE "bruh seems …
2 R_PBtJhKNv0JB10GJ Masculine Straight 26-40 FALSE TRUE "bruh is gen…
3 R_305B45bvuFeKZUx Feminine Bisexual 18-25 FALSE TRUE "i don’t us…
4 R_305B45bvuFeKZUx Feminine Bisexual 18-25 FALSE TRUE "i think it …
5 R_vjZzVyscBlf29AB Masculine cis 41-55 FALSE TRUE "bruh is mor…
6 R_1IL6Ojmmkco93Hi Feminine bi 41-55 FALSE TRUE "i don't rea…
# load up stopwords
snowball <- stopwords(language = "en", source = "snowball", simplify = TRUE)
nltk <- stopwords(language = "en", source = "nltk", simplify = TRUE)
smart <- stopwords(language = "en", source = "smart", simplify = TRUE)
# clean up -- sents containing 'bro'
bro_final <- bro_clean %>%
unnest_tokens(TOKEN, `SENT-TOK2`) %>%
filter(!(TOKEN %in% snowball | TOKEN %in% nltk | TOKEN %in% smart |
TOKEN %in% c("bro", "bros", "bruh", "don’t", 'i’m', 'it’s',
'people', "person", 'term', "address", "expression", "feel", "feels")))
# clean up -- sents containing 'bruh'
bruh_final <- bruh_clean %>%
unnest_tokens(TOKEN, `SENT-TOK2`) %>%
filter(!(TOKEN %in% snowball | TOKEN %in% nltk | TOKEN %in% smart |
TOKEN %in% c("bro", "bros", "bruh", "don’t", 'i’m', 'it’s',
'people', "person", 'term', "address", "expression", "feel", "feels")))# `bro` plot
bro_final %>%
count(TOKEN, sort = TRUE, name = "Count") %>%
mutate(TOKEN = reorder(TOKEN, Count)) %>%
head(20) %>%
ggplot(aes(x = Count, y = TOKEN)) +
geom_col() +
labs(title = "Top 20 Tokens Used to Describe 'Bro'")
#`bruh` plot
bruh_final %>%
count(TOKEN, sort = TRUE, name = "Count") %>%
mutate(TOKEN = reorder(TOKEN, Count)) %>%
head(20) %>%
ggplot(aes(x = Count, y = TOKEN)) +
geom_col() +
labs(title = "Top 20 Tokens Used to Describe 'Bruh'")