distribution of zodiac signs in US population

zs_dist.R

library(tidyverse)
library(lubridate)

# modified from https://stackoverflow.com/questions/48026462/writing-r-function-using-lubridate-package-that-takes-dateex-august-14-and
zodiac_sign <- function(input)
{
    zodiac_sign_df <- data.frame(Month = c("March", "April", "May", "June", "July", "August", "September", "October", "November", "December", "January", "February"),
                                 Zodiac_Separation_Dt = c(21, 20, 21, 21, 23, 23, 23, 23, 22, 22, 20, 19),
                                 LowerSideZS = c("Pisces", "Aries","Taurus","Gemini","Cancer","Leo","Virgo","Libra","Scorpio","Sagittarius","Capricorn","Aquarius"),
                                 UpperSideZS = c("Aries","Taurus","Gemini","Cancer","Leo","Virgo","Libra","Scorpio","Sagittarius","Capricorn","Aquarius", "Pisces"),
                                 stringsAsFactors = FALSE )
    val_df <- zodiac_sign_df[match(months(input), zodiac_sign_df$Month), ]
    ifelse( day(input) >= val_df$Zodiac_Separation_Dt, 
            val_df$UpperSideZS, val_df$LowerSideZS)
}

# data from fivethirtyeight - https://github.com/fivethirtyeight/data/tree/master/births
SSA_data <- "https://raw.githubusercontent.com/fivethirtyeight/data/master/births/US_births_2000-2014_SSA.csv"
CDC_NCHS_data <- "https://raw.githubusercontent.com/fivethirtyeight/data/master/births/US_births_1994-2003_CDC_NCHS.csv"

dat <- read.csv(SSA_data)

dat %>%
    mutate(date = ISOdate(year, month, date_of_month), 
           zs = zodiac_sign(date)) %>%
    count(zs, wt = births) %>%
    mutate(prop = n / sum(n)) %>%
    arrange(desc(prop))
    
 # very different from http://faculty.tamucc.edu/sfriday/wordpress/?p=1317#:~:text=Most%20Common%20to%20Least%20Common%20Zodiac%20Signs%20,%200.090%20%25%20%208%20more%20rows%20