Counting 0s 1s and 2s in multiple column in r - r

Question

My data looks like this:
structure(list(did = c(209L, 209L, 206L, 206L, 206L, 206L, 206L,
206L, 206L, 206L, 206L, 209L, 206L, 206L, 207L, 207L, 207L, 207L,
209L, 209L), hhid = c(5668, 5595, 4724, 4756, 4856, 4730, 4757,
6320, 4758, 6319, 6311, 5477, 6322, 6317, 134, 178, 238, 179,
5865, 5875), bc = c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 0L, 1L, 1L, 1L,
1L, 0L, 1L, 1L, 1L, 1L, 1L, 0L, 0L), rc = c(1L, 1L, 1L, 1L, 1L,
1L, 1L, 0L, 1L, 1L, 1L, 1L, 0L, 1L, 1L, 1L, 1L, 1L, 1L, 1L),
oap = c(2L, 2L, 0L, 0L, 0L, 0L, 0L, 2L, 0L, 2L, 2L, 0L, 2L,
2L, 2L, 2L, 2L, 2L, 0L, 0L)), row.names = c(NA, 20L), class = "data.frame")
hhid is unique for each row. For the remaining rows it consist of 0s and 1s in some columns and 0s 1s and 2s in other columns.
The output column required is like this:
did hh_count bc_0 bc_1 bc_2 rc_0 rc_1 rc_2 oap_0 oap_1 oap_2
where did will be unique.hh_count will be count of each hhid associated with did.
bc_0, bc_1 and bc_1 will be breakup of column bc and it will represent count of 0s 1s and 2s in bc.Simmilarily for rc_0,rc_1and rc_2 and oap_0,oap_1 and oap_2.So counting of 0s 1s and 2s is required

Answer 1

With counts of 3 specific values, writing the functions manually seems reasonable. If you need specific counts of more distinct values we could come up with a better way to generalize - probably converting your data to long format, summarizing, and then going back to wide.
library(dplyr) # across() requires dplyr version 1.0 or higher
dd %>% # (calling your data dd)
group_by(did) %>%
summarize(
hh_count = n_distinct(hhid),
across(c(bc, rc, oap),
.fns = list("0" = ~sum(. == 0), "1" = ~sum(. == 1), "2" = ~sum(. == 2)),
.names = "{.col}_{.fn}" # this is the default, but I show it explicitly
)
)
# # A tibble: 3 x 11
# did hh_count bc_0 bc_1 bc_2 rc_0 rc_1 rc_2 oap_0 oap_1 oap_2
# <int> <int> <int> <int> <int> <int> <int> <int> <int> <int> <int>
# 1 206 11 2 9 0 2 9 0 6 0 5
# 2 207 4 0 4 0 0 4 0 0 0 4
# 3 209 5 2 3 0 0 5 0 3 0 2