dplyr: Use a custom function in summarize() after group_by()

Question

How can we use a custom function after group_by()? I checked similar posts (1, 2, and 3), but my current code returns the same values for all groups.

> data
   village     A     Z      Y 
     <chr> <int> <int>   <dbl> 
 1       a     1     1   500     
 2       a     1     1   400     
 3       a     1     0   800  
 4       b     1     0   300  
 5       b     1     1   700  

z <- 1
data %>%
    group_by(village) %>%
    summarize(Y_village = Y_hat_village(., z))

Y_hat_village <- function(data_village, z){
    # Calculate the mean for a specific z in a village
    data_z <- data_village %>% filter(Z==get("z"))
    return(mean(data_z$Y))
}

I want to have (500 + 400)/2 = 450 for village "a" and 700 for village "b".

Answer 1

It's easier to understand if you start by writing it without an extra function. In that case it would be:

df %>%
  group_by(village) %>%
  summarize(Y_village = mean(Y[Z == z]))

## A tibble: 2 x 2
#  village Y_village
#  <fct>       <dbl>
#1 a            450.
#2 b            700.

Hence, your function should be something like

Y_hat_village <- function(Ycol, Zcol, z){
  mean(Ycol[Zcol == z])
}

And then using it:

df %>%
  group_by(village) %>%
  summarize(Y_village = Y_hat_village(Y, Z, z))

Note that the function I wrote only deals with atomic vectors which you can supply directly from within summarise. You don't need to supply the whole data.frame into it.