Using babynames package to illustrate complex data sharing

DescribeDataSharing.R

library(babynames)
library(tidyverse)

set.seed(8746)

snames <- sample(babynames$name, 14)
unames <- snames[11:14]
snames <- snames[1:10]

Agency1 <- data_frame(Person = c(sample(snames, 5), unames[1]),
                  Agency1 = "X") %>% 
  arrange(Person)

Agency2 <- data_frame(Person = c(sample(snames, 6), unames[2]),
                  Agency2 = "X") %>% 
  arrange(Person)

Agency3 <- data_frame(Person = c(sample(snames, 6), unames[3]),
                  Agency3 = "X") %>% 
  arrange(Person)

Agency4 <- data_frame(Person = c(sample(snames, 8), unames[4]),
                  Agency4 = "X") %>% 
  arrange(Person)

Agency1_Agency2 <- inner_join(Agency1, Agency2)

Agency1_Agency3 <- inner_join(Agency1, Agency3)

Agency1_Agency4 <- inner_join(Agency1, Agency4)

Agency2_Agency3 <- left_join(Agency2, Agency3)

Agency4_Agency3 <- left_join(Agency4, Agency3)

Agency1_Agency2_Agency3 <- inner_join(Agency1, Agency2_Agency3)

Agency1_Agency4_Agency3 <- inner_join(Agency1, Agency4_Agency3)

option_0 <-  full_join(Agency1_Agency2, Agency1_Agency4) %>% full_join(Agency1_Agency3) %>% select(Person, Agency1, Agency2, Agency4, Agency3)

option_1 <- full_join(Agency1_Agency2_Agency3, Agency1_Agency4_Agency3) %>% select(Person, Agency1, Agency2, Agency4, Agency3)

option_2 <- full_join(Agency1_Agency2_Agency3, Agency1_Agency4) %>% select(Person, Agency1, Agency2, Agency4, Agency3)