Women in Technology StackOverflow Sruvey

gistfile1.txt

library(stacksurveyr)
library(dplyr)
library(tidyr)
library(purrr)
library(stringr)
library(ggplot2)
library(forcats)

df <- stack_survey %>%
  select(respondent_id, gender, tech_do) %>% 
  filter(!is.na(gender) & !is.na(tech_do)) 

tech <- str_split(df$tech_do, ";")

reshape_data <- function(str, gender){
  data_frame(gender = gender, tech = str)
}  

df <- map2_df(tech, df$gender, ~reshape_data(.x, .y))

gender_dist <- df %>% 
  group_by(tech, gender) %>% 
  tally() %>% 
  mutate(percent_total = n / sum(n))

popular_female <- gender_dist %>% 
  filter(gender == 'Female') %>%
  filter(n > 10) %>% 
  arrange(desc(percent_total))

popular_female %>% 
  filter(tech %in% popular_female$tech) %>% 
  .[1:20, ] %>% 
  ggplot( aes(x = percent_total, 
              y = fct_reorder(tech, as.numeric(percent_total))))+
  geom_point() +
  xlab("Percentage of Stackoverflow Survey Responses") + 
  ylab("Total")+
  ggtitle("Technologies with more than 10 women responses")