Mental health survey
library(tidyverse)
dat<-read_csv("C:/Users/Dr Mohsen/Documents/R/coursera/data_visualization_capestone/survey.csv")
Relation between respondent’s age and seeking mental treatment
table(dat$treatment, useNA = "ifany")
table(dat$Age, useNA = "ifany")
# remove outliers with negative or more than 100 years
dat2<-dat %>% filter(Age > 0 & Age < 100)
ggplot(data = dat2, aes(x = treatment, y = Age, col = treatment))+
geom_boxplot(outlier.shape = NA)+
geom_jitter()+
labs(title = "Relation between respondent's age and seeking mental treatment",
x = "Seeking mental treatment", col = "Seeking mental treatment",
y ="Age (years)")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
axis.text = element_text(face = "bold"))
Relation between Respondent’s country and seeking mental
treatment
library(plotly)
table(dat$Country, useNA = "ifany")
df<-dat %>% group_by(Country) %>% count(treatment) %>%
dplyr::filter(sum(n) > 10) %>%
dplyr::mutate(percentage = round((n/sum(n))*100)) %>%
dplyr::filter(treatment=="Yes") %>% ungroup() %>%
dplyr::mutate(Country = reorder(Country, percentage))
p<-ggplot(data = df, aes(x=percentage,y= Country))+
geom_point(size=3)+
geom_segment(aes(x = 0, xend = percentage, y = Country,
yend = Country))+
scale_x_continuous(expand = c(0,1))+
labs(x="Percentage seeking treatment",y="")+
theme_classic()
ggplotly(p) %>%
layout(title = list(text = paste0("Relation between respondent's country and percentage seeking mental treatment",
'<br>',
'<sup>',
'For countries with more than 10 respondents','</sup>'), font = list(size = 14)))
Relation between Respondent’s U.S. state and seeking mental
treatment
library(maps)
df<-data.frame(state = state.name, state_abb = state.abb)
dat2<-dat %>% left_join(df, by = c("state" = "state_abb"))
my_state_map <- map_data("state")
df2<-dat2 %>% group_by(state.y) %>% count(treatment) %>%
dplyr::filter(sum(n) > 1) %>%
dplyr::mutate(percentage = round((n/sum(n))*100)) %>%
dplyr::filter(treatment=="Yes") %>% drop_na() %>%
dplyr::mutate(state.y = str_to_lower(state.y))
df3<-left_join(my_state_map,df2,by=c("region"= "state.y"))
p<-ggplot(data = df3, mapping = aes(x= long, y= lat,
group = group,
fill= percentage, col = ""))+
geom_polygon()+
scale_fill_gradient2(low="brown3", mid="cornsilk1", high="turquoise4",
limits=c(20, 100), na.value="black") +
scale_colour_manual(values= NA) +
guides(colour=guide_legend("No data",
override.aes=list(colour="black")))+
labs(title = "Relation between respondent's U.S. state and percentage seeking mental treatment",
subtitle = "For states with more than 1 respondent",
fill = "Percentage seeking mental treatment")+
theme_void()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 10, hjust = 0.5))
p
Relation between self employed and seeking treatment
table(dat$self_employed, useNA = "ifany")
df<-dat %>% drop_na(self_employed) %>% group_by(self_employed) %>%
count(treatment) %>%
mutate(percentage = round((n/sum(n))*100))
ggplot(data = dat %>% drop_na(self_employed),
aes(x = self_employed, fill = treatment))+
geom_bar(position = "fill", width = 0.5)+
geom_text(data = df, aes(x = self_employed, y = n,
label = paste(percentage,"%")),
position = position_fill(vjust = 0.5))+
scale_y_continuous(labels = scales::percent_format())+
labs(title = "Relation between self employement and seeking mental treatment",
x = "Self employement", fill = "Seeking mental treatment",
y = "Percentage")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
axis.text = element_text(face = "bold"))
Relation between family history and seeking treatment
table(dat$family_history, useNA = "ifany")
df<-dat %>% drop_na(family_history) %>% group_by(family_history) %>%
count(treatment) %>%
mutate(percentage = round((n/sum(n))*100))
ggplot(data = dat %>% drop_na(family_history),
aes(x = family_history, fill = treatment))+
geom_bar(position = "fill", width = 0.5)+
geom_text(data = df, aes(x = family_history, y = n,
label = paste(percentage,"%")),
position = position_fill(vjust = 0.5))+
scale_y_continuous(labels = scales::percent_format())+
labs(title = "Relation between Family history of mental illness \n and seeking mental treatment",
x = "Family history of mental illness", fill = "Seeking mental treatment",
y = "Percentage")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
axis.text = element_text(face = "bold"))
Relation between employees number in the company and seeking mental
treatment
table(dat$no_employees, useNA = "ifany")
df<-dat %>% drop_na(no_employees) %>% group_by(no_employees) %>%
count(treatment) %>%
mutate(percentage = round((n/sum(n))*100))
ggplot(data = dat %>% drop_na(no_employees) %>%
dplyr::mutate(no_employees = fct_relevel(no_employees,
"1-5", "6-25",
"26-100", "100-500",
"500-1000",
"More than 1000")),
aes(x = no_employees, fill = treatment))+
geom_bar(position = "fill", width = 0.5)+
geom_text(data = df, aes(x = no_employees, y = n,
label = paste(percentage,"%")),
position = position_fill(vjust = 0.5))+
scale_y_continuous(labels = scales::percent_format())+
labs(title = "Relation between employees number in the company \n and seeking mental treatment",
x = "Employees number", fill = "Seeking mental treatment",
y = "Percentage")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
axis.text = element_text(face = "bold"))+
coord_flip()
Relation between remote work and seeking treatment
table(dat$remote_work, useNA = "ifany")
df<-dat %>% drop_na(remote_work) %>% group_by(remote_work) %>%
count(treatment) %>%
mutate(percentage = round((n/sum(n))*100))
ggplot(data = dat %>% drop_na(remote_work),
aes(x = remote_work, fill = treatment))+
geom_bar(position = "fill", width = 0.5)+
geom_text(data = df, aes(x = remote_work, y = n,
label = paste(percentage,"%")),
position = position_fill(vjust = 0.5))+
scale_y_continuous(labels = scales::percent_format())+
labs(title = "Relation between working remotely and seeking mental treatment",
x = "Remote work", fill = "Seeking mental treatment",
y = "Percentage")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
axis.text = element_text(face = "bold"))
Relation between working in a tech company and seeking mental
treatment
table(dat$tech_company, useNA = "ifany")
df<-dat %>% drop_na(tech_company) %>% group_by(tech_company) %>%
count(treatment) %>%
mutate(percentage = round((n/sum(n))*100))
ggplot(data = dat %>% drop_na(tech_company),
aes(x = tech_company, fill = treatment))+
geom_bar(position = "fill", width = 0.5)+
geom_text(data = df, aes(x = tech_company, y = n,
label = paste(percentage,"%")),
position = position_fill(vjust = 0.5))+
scale_y_continuous(labels = scales::percent_format())+
labs(title = "Relation between working in a tech company and seeking mental treatment",
x = "Working in a tech company", fill = "Seeking mental treatment",
y = "Percentage")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
axis.text = element_text(face = "bold"))
Relation between Country and ease of mental medical leave
table(dat$leave, useNA = "ifany")
df<-dat %>% group_by(Country) %>%
count(leave) %>% filter(sum(n)>10) %>% filter(!leave=="Don't know") %>%
mutate(percentage = round((n/sum(n))*100)) %>% ungroup() %>%
dplyr::mutate(leave = fct_relevel(leave,
"Very easy", "Somewhat easy",
"Somewhat difficult",
"Very difficult"))
ggplot(data = dat %>% filter(Country %in% df$Country) %>%
filter(!leave=="Don't know") %>%
dplyr::mutate(leave = fct_relevel(leave,
"Very easy", "Somewhat easy",
"Somewhat difficult",
"Very difficult")),
aes(x = Country, fill = leave))+
geom_bar(position = "fill", width = 0.5)+
geom_text(data = df, aes(x = Country, y = n,
label = paste(percentage,"%")),
position = position_fill(vjust = 0.5))+
scale_y_continuous(labels = scales::percent_format())+
labs(title = "Relation between medical leave for a mental health condition \n and respondent's country",
subtitle = "Countries with more than 10 respondents",
x = "", fill = "Ease of medical mental leave",
y = "Percentage")+
theme_classic()+
theme(plot.title = element_text(size = 12, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 10, hjust = 0.5),
axis.text = element_text(face = "bold"))+
coord_flip()
