Aggregating numbers per factor level with different lengths

Question

I think the easiest way to explain what I want is to give you a simple example.

So this is my mock data set:

example <- data.frame(location = c(rep('Location A', 5), rep('Location B', 4), rep('Location C', 7)), 
                  factor_lvl = as.factor(c(paste('level', c(1:5), sep = ' '), 
                                           paste('level', c(1:4), sep = ' '),
                                           paste('level', c(1:7), sep = ' '))),
                  no_answers = floor(runif(16, min=0, max=20) ))

which therefore looks like this

     location factor_lvl no_answers
1  Location A    level 1          1
2  Location A    level 2         13
3  Location A    level 3          4
4  Location A    level 4          8
5  Location A    level 5          6
6  Location B    level 1         13
7  Location B    level 2         17
8  Location B    level 3         15
9  Location B    level 4          7
10 Location C    level 1          5
11 Location C    level 2          8
12 Location C    level 3          1
13 Location C    level 4         19
14 Location C    level 5         13
15 Location C    level 6         18
16 Location C    level 7          0

What i want is aggregate the number of answers per each location and repeat this number until the new location. For example, Location A has 5 factor levels, total number of answers is 32, so I want the new column to start with 5 32s in a row etc.

To be more clear, the desired output is like this:

wanted_result <- cbind(example, total_answers = c(rep(32,5), rep(52, 4), rep(64, 7)  ) )


   location      factor_lvl no_answers total_answers
1  Location A    level 1          7            32
2  Location A    level 2         16            32
3  Location A    level 3         14            32
4  Location A    level 4          8            32
5  Location A    level 5          7            32
6  Location B    level 1          0            52
7  Location B    level 2          7            52
8  Location B    level 3          5            52
9  Location B    level 4          2            52
10 Location C    level 1         13            64
11 Location C    level 2          3            64
12 Location C    level 3         10            64
13 Location C    level 4         19            64
14 Location C    level 5         13            64
15 Location C    level 6          7            64
16 Location C    level 7          2            64

`with(example, ave(no_answers, location, FUN = sum))` – Ronak Shah Feb 22 '19 at 12:51 — Ronak Shah, Feb 22 '19 at 12:51

score 1 · Accepted Answer · answered Feb 22 '19 at 12:50

1

Do you just need:

library(dplyr)

example %>%
  group_by(location) %>%
  mutate(total_answers = sum(no_answers))

answered Feb 22 '19 at 12:50

arg0naut91

14,574
2
17
38

Yes, exactly, thanks. It's so simple I'm actually ashamed I had to look here for an answer. – Astronaut Feb 22 '19 at 13:06
You're welcome, we've all been there! – arg0naut91 Feb 22 '19 at 13:09

Aggregating numbers per factor level with different lengths

1 Answers1