expand a 2 by 2 contingency table-CodePudding

I have loaded data in the following format:

Gender	Yes	No
Male	2	1
Female	1	2

I would like to expand it to:

Gender	Result
Male	Yes
Male	Yes
Male	No
Female	Yes
Female	No
Female	No

I've tried using the expand.table function in epitools but have not succeeded.

CodePudding user response：

Using tidyr, you can pivot the data into long format, then uncount it.

Assuming your data is called df, you could do that as follows:

library(tidyr)

pivot_longer(df, -Gender, names_to = 'Result') %>% 
  uncount(value)
#> # A tibble: 6 x 2
#>   Gender Result
#>   <chr>  <chr> 
#> 1 Male   Yes   
#> 2 Male   Yes   
#> 3 Male   No    
#> 4 Female Yes   
#> 5 Female No    
#> 6 Female No

Data used, taken from question and put into reproducible format

df <- structure(list(Gender = c("Male", "Female"), Yes = 2:1, No = 1:2), 
                class = "data.frame", row.names = c(NA, -2L))

CodePudding user response：

Here is a data.table solution:

df1 <- data.frame(Gender = c("Male", "Female"), 
                  Yes = c(2,1), No = c(1,2))

library(data.table)

melt(setDT(df1), id.vars = "Gender")[,.SD[rep(.I, value)]][,value:=NULL][]

#>    Gender variable
#> 1:   Male      Yes
#> 2:   Male      Yes
#> 3: Female      Yes
#> 4:   Male       No
#> 5: Female       No
#> 6: Female       No

CodePudding user response：

Here is another solution based on data.table package:

library(data.table)

setDT(df)[, .(result = rep.int(c("Yes", "No"), c(Yes, No))), by=Gender]

#    Gender result
# 1:   Male    Yes
# 2:   Male    Yes
# 3:   Male     No
# 4: Female    Yes
# 5: Female     No
# 6: Female     No