r - Average the periods 1-3 and 4-6 for each treatment and id for each variable and get the data in a new data frame

Question

Here I would like to average the periods 1-3 and 4-6 for each treatment and id for each variable and get the data in a new data frame. Does anyone know how I accomplish this?

set.seed(1)
id <- rep(1:2,each=6)
trt <- c("A","A","A","B", "B", "B","A","A","A","B", "B", "B")
period <- rep(1:6,2)
pointA <- sample(1:10,12, replace=TRUE)
pointB<- sample(1:10,12, replace=TRUE)
pointC<- sample(1:10,12, replace=TRUE)
df <- data.frame(id,trt,period,pointA, pointB,pointC)
head(df)

   id trt period pointA pointB pointC
1   1   A      1      3      7      3
2   1   A      2      4      4      4
3   1   A      3      6      8      1
4   1   B      4     10      5      4
5   1   B      5      3      8      9
6   1   B      6      9     10      4
7   2   A      1     10      4      5
8   2   A      2      7      8      6
9   2   A      3      7     10      5
10  2   B      4      1      3      2
11  2   B      5      3      7      9
12  2   B      6      2      2      7

I would like it to look like this:

  id trt Period pointA pointB pointC
1  1   A    123     13     19      8
2  1   B    456     21     23     17
3  2   A    456     24     22     16
4  2   B    123      6     12     18

标签： r

Answer 1

使用dplyr您可以使用适当的组创建一个新变量，然后将其用作 group_by。例如

library(dplyr)
df %>% 
  mutate(period_class = case_when(
    period %in% c(1,2,3)~"123",
    period %in% c(4,5,6)~"456")
  ) %>% 
  select(-period) %>% 
  group_by(id, trt, period_class) %>% 
  summarize_all(mean) # though you seem to have used `sum` in your example