Similar to Countifs and Sumifs in R

Question

I should preface this... I cannot use dplyr. It is just not installing in my version of R.

How would I do an action similar to a countifs or sumifs function in R?

   P1 | P2 | Match | Same | count_of_friends

   M  | F  | FALSE | FALSE| 6
   M  | M  | TRUE  | TRUE | 7
   F  | M  | FALSE | FALSE| 10
   F  | F  | TRUE  | FALSE| 2

I would essentially be looking for something similar to EXCEL's

SUMIFS(Match == Same; count_of_friends)

I want to find the sum of friends if both individuals are of the same gender, or the sum of the counts of friends if the P1 is a Female.

I then also want to find out how to count only the instances where the count of friends is over 5, etc.

How would you do this in R?

What's your expected output? Not everyone is familiar with functions `countifs` and `sumifs` — pogibas, Mar 04 '18 at 08:51
Yesterday too you asked a similar question without much clarity — akrun, Mar 04 '18 at 08:57
I'm sorry - I'm not sure how it is unclear. It is not exactly one specific question so much as a general. How would I sum up the integers in a column for only the instances where a condition for another column is met? And how would this be done on count? — Kimberly, Mar 04 '18 at 09:01
@Kimberly you have a total of 9 questions in which you have not accepted any answers on stack overflow. All of the questions are of poor quality and without a reproducible example. Please read SO guidelines:[1](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example), [2](https://stackoverflow.com/help/how-to-ask), [3](https://stackoverflow.com/help/someone-answers). If this persists you are risking your further questions being ignored. — missuse, Mar 04 '18 at 09:14

score 0 · Accepted Answer · answered Mar 04 '18 at 09:02

Here is an approach in base R:

first question, subset the data frame according to a logical vector P1 == P2 and sum the values in column 5

sum(df[with(df, P1 == P2), 5])
#output
9

second question, subset the data frame according to a logical vector count_of_friends > 5 and check the number of rows of the resulting data frame:

nrow(df[with(df, count_of_friends > 5),])
#output
3

data:

> dput(df)
structure(list(P1 = structure(c(2L, 2L, 1L, 1L), .Label = c("F", 
"M"), class = "factor"), P2 = structure(c(1L, 2L, 2L, 1L), .Label = c("F", 
"M"), class = "factor"), Match = c(FALSE, TRUE, FALSE, TRUE), 
    Same = c(FALSE, TRUE, FALSE, FALSE), count_of_friends = c(6, 
    7, 10, 2)), .Names = c("P1", "P2", "Match", "Same", "count_of_friends"
), row.names = c(NA, -4L), class = "data.frame")

score 0 · Answer 2 · answered Mar 04 '18 at 09:09

We can use dplyr to filter rows that have 'P1' value equal to 'P2' and then summarise the 'count_of_friends' by taking the sum

library(dplyr)
df %>%
   filter(P1 == P2) %>% 
   summarise(Sum = sum(count_of_friends))
#   Sum
#1   9

For the second part, we do the filter on the 'count_of_friends' and get the nrow

df %>%
   filter(count_of_friends > 5) %>%
   nrow
#[1] 3

Similar to Countifs and Sumifs in R

2 Answers2