Archaeology job ads from Archaeology Jobs Wiki

gistfile1.txt

library(tidyverse)

base_url <- "http://academicjobs.wikia.com/wiki/Archaeology_Jobs_"

# starts at 2010-2011

years <- map_chr(2010:2019, ~str_glue('{.x}-{.x +1}'))
# though it seems to start at 2007-8: https://academicjobs.fandom.com/wiki/Archaeology_07-08

urls_for_each_year <- str_glue('{base_url}{years}')

library(rvest)

# tenure but not non-tenure

#------------------------------------------

# 2010-2011 has no table
urls_for_each_year[1] %>% 
  read_html() %>% 
  # html_nodes('.mw-content-text') %>% 
  html_nodes('.mw-headline') %>% 
  html_text()

#------------------------------------------
# table first appears in 2011-2012
urls_for_each_year[2] %>% 
  read_html() %>%     
  html_node('table , td') %>% 
  html_table()

# but headings are not systematic
urls_for_each_year[2] %>% 
  read_html() %>% 
  #   html_nodes('.mw-content-text') %>% 
  html_nodes('.mw-headline') %>% 
  html_text()

#------------------------------------------
# table first appears in 2012-2013
urls_for_each_year[3] %>% 
  read_html() %>% 
  html_node('table , td') %>% 
  html_table()

urls_for_each_year[3] %>% 
  read_html() %>% 
  # html_nodes('.mw-content-text') %>% 
  html_nodes('.mw-headline') %>% 
  html_text()

#-------------------------------------
# all years
urls_for_each_year_headers <- 
map(urls_for_each_year,
    ~.x %>% 
      read_html() %>% 
      html_nodes('.mw-headline') %>% 
      html_text())

# what are the different sections?
tt_sections <- c("TENURE-TRACK POSITIONS",
                "TENURE-TRACK OR TENURED / FULL-TIME POSITIONS",
               "Tenure-Track or Tenured / Full-time Position",
               "ASSISTANT PROFESSOR OR OPEN RANK",
               "TENURE TRACK ASSISTANT PROFESSOR OR OPEN RANK",
               "TENURED ASSOCIATE OR FULL PROFESSOR",
               "ASSOCIATE OR FULL PROFESSOR")

non_tt_sections <- c("NON-TENURE-TRACK POSITIONS", 
                     "VISITING POSITIONS / Limited-Term Appointments / Postdocs", 
                     "Visiting Positions / Limited-Term Appointments / Postdocs",
                     "VISITING POSITIONS / LIMITED TERM APPOINTMENTS / POSTDOCS",   
                     "VISITING POSITIONS / LIMITED-TERM APPOINTMENTS / POSTDOCS / PART-TIME POSITIONS",
                     "VISITING POSITIONS")

end_non_tt_sections <- c("DISCUSSION, RUMORS AND SPECULATION",   
                         "DISCUSSION, RUMORS, SPECULATION",
                         "General Discussion, Rumors, and Speculation" )

# what to do about this?
# "COMPLETED SEARCHES"

library(stringr)
get_counts_tt_non_tt <- function(x){
  
# x <- urls_for_each_year_headers[[9]]
tt_start      <- which(str_detect(x, 
                                  paste(tt_sections, 
                                        collapse = "|")))[1] 
non_tt_start   <- which(str_detect(x, 
                                   paste(non_tt_sections, 
                                         collapse = "|")))[1] 
non_tt_end    <- which(str_detect(x, 
                                  paste(end_non_tt_sections, 
                                        collapse = "|")))[1] 
n_tt_jobs     <- length( x[(tt_start + 1) : (non_tt_start - 1) ] )
n_non_tt_jobs <- length( x[(non_tt_start + 1) : (non_tt_end  - 1) ] )

return(list(n_tt_jobs = n_tt_jobs,
            n_non_tt_jobs = n_non_tt_jobs))
}

map_df(urls_for_each_year_headers,
    get_counts_tt_non_tt) %>% 
  mutate(ratio = n_tt_jobs / n_non_tt_jobs) %>% 
  mutate(year = str_replace(str_sub( urls_for_each_year, -9),
                            "-", "-\n")) %>% 
  ggplot() +
  aes(year, 
      group = 1,
      ratio) +
  geom_line() +
  geom_hline(yintercept = 1,
             colour = "red") +
  labs(y = "Ratio of tenure-track to non-tenure track\njobs in Archaeology",
       x = "") +
  theme_bw(base_size = 16)
  

##---------------------------------------------------

library(googlesheets4)
library(tidyverse)

jobdata<-read_sheet("https://docs.google.com/spreadsheets/d/1Jwe3UqJyedrV-QWlwR_44__t4xBVrCfxGyhXdi3E0sg/edit#gid=1686084773")

integer_breaks <- function(n = 5, ...) {
  fxn <- function(x) {
    breaks <- floor(pretty(x, n, ...))
    names(breaks) <- attr(breaks, "labels")
    breaks
  }
  return(fxn)
}

jobdata %>% 
  select(starts_with("Documents requested")) %>% 
  pivot_longer(everything()) %>% 
  group_by(name, value) %>% 
  tally() %>% 
  mutate(value = case_when(
    value == "Not requested in the job ad" ~ 0,
    value == "One" ~ 1,
    value == "Two (e.g. two syllabi)" ~ 2,
    value == "Three" ~ 3,
    .default = 0
  )) %>% 
  ggplot() +
  aes(value,
      n) +
  geom_col() +
  facet_wrap( ~ name) +
  scale_x_continuous(label = integer_breaks(),
                     limits = c(-1, 4))

# do the requirements differ for associate positions 
jobdata %>% 
  mutate(position_title = case_when(
    str_detect(`Title of position`, "Associate") ~ "Associate",
    str_detect(`Title of position`, "Assistant") ~ "Assistant")) %>% 
  select(position_title,
         starts_with("Documents requested")) %>% 
  pivot_longer(-position_title) %>% 
  mutate(value = case_when(
    value == "Not requested in the job ad" ~ 0,
    value == "One" ~ 1,
    value == "Two (e.g. two syllabi)" ~ 2,
    value == "Three" ~ 3,
    .default = 0
  )) %>% 
  filter(!is.na(position_title)) %>% 
  ggplot() +
  aes(position_title,
      value
      ) +
    geom_boxplot() +
  facet_wrap( ~ name)