I recently received an e-mail from the American Statistical Association (ASA) about the Helen Walker Society. The e-mail listed the 14 “distinguished women leaders who have held the office of ASA president” going back to Helen Walker in 1944.
Really? Fourteen? All the way back to 1944? Wow.
I thought it would be interesting to visualize the percentage of female ASA presidents over time. I found a listing of ASA presidents on Wikipedia and used the data to create a timeline.
library(rvest)
library(plyr)
library(DT)
library(zoo)
library(tidyverse)
pres1 <- html("https://en.wikipedia.org/wiki/List_of_presidents_of_the_American_Statistical_Association") %>%
html_node("#mw-content-text") %>%
html_text() %>%
strsplit("\n")
pres2 <- pres1[[1]]
pres3 <- pres2[grepl("^\\d{4}", unlist(pres2))]
df <- data.frame(
Year = as.numeric(substring(pres3, 1, 4)),
President = trimws(gsub("\\d|-", "", pres3))
) %>%
mutate(
Number = seq_along(President),
Sex = ifelse(Year %in% c(1944, 1952, 1956, 1980, 1987, 1989, 1992, 1996,
2006, 2007, 2009, 2011, 2013, 2016, 2018), "Female", "Male")
) %>%
select(Number, Year, President, Sex)
datatable(df, rownames=FALSE,
caption="Presidents of the American Statistical Association, listed by their first year of service")
I then calculated the rolling 10-year percentage of ASA presidents that were female and plotted the results.
df2 <- left_join(data.frame(Year=min(df$Year):max(df$Year)), df, by="Year")
df3 <- colwise(na.locf, fromLast=TRUE)(df2) %>%
mutate(
mean10 = 100*rollmean(Sex=="Female", 10, align="right", fill=NA)
)
ggplot(df3, aes(Year, mean10)) +
geom_hline(yintercept=50, linetype=2) +
geom_line(size=1, color="orange") +
ggtitle("Female ASA Presidents") +
labs(y = "Female presidents (% of past 10 y)") +
theme_bw(base_size=16)
There were three female ASA presidents between 1944 and 1956, followed by a 23-year period during which all the ASA presidents were male (1957-1979). Since then, the percentage of female ASA presidents has risen, reaching 50% for the first time during the 10-year period from 2004 to 2013. Excellent.