Wide to long dataframe with many within and between subject variables

user_16939164 0
Viewed 83

I'm trying to build a data set with a long type of structure, with 2 between-subject variables and 2 within-subject variables from an excel table. The current dataset structure is the following:

> str(Subset_0)         
'data.frame':   54 obs. of  11 variables:
 $ Subject : num  1 2 3 4 5 6 7 8 9 10 ...
 $ BETWEEN1: num  1 1 1 2 2 2 2 1 1 2 ...
 $ BETWEEN2: num  1 1 2 2 2 2 1 1 1 1 ...
 $ A_x1    : num  5 1 3 1 0 6 1 2 7 1 ...
 $ B_x2    : num  5 1 3 0 3 0 0 2 6 1 ...
 $ C_y1    : num  6 9 9 2 2 4 2 2 6 0 ...
 $ D_y2    : num  6 15 4 1 2 4 3 1 3 0 ...
 $ K_x1    : num  5 1 3 1 0 6 1 2 7 1 ...
 $ L_x2    : num  5 1 3 0 3 0 0 2 6 1 ...
 $ M_y1    : num  6 9 9 2 2 4 2 2 6 14 ...
 $ N_y2    : num  3 1 0 4 0 5 6 5 17 21 ...

data file from dput:

structure(list(Subject = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 
12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 
28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 
44, 45, 46, 47, 48, 49, 50, 51, 52, 54, 55), BETWEEN1 = c(1, 
1, 1, 2, 2, 2, 2, 1, 1, 2, 2, 2, 2, 1, 2, 1, 2, 1, 1, 2, 2, 1, 
2, 1, 2, 1, 1, 2, 1, 1, 2, 1, 1, 2, 1, 2, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA), BETWEEN2 = c(1, 
1, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 
1, 1, 2, 2, 1, 1, 2, 2, 1, 1, 2, 2, 2, 2, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA), A_x1 = c(5, 
1, 3, 1, 0, 6, 1, 2, 7, 1, 1, 0, 0, 2, 0, 8, NA, NA, NA, NA, 
14, 23, 19, 10, 9, 10, 11, 14, 16, 8, 24, 17, 8, 22, 14, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA), B_x2 = c(5, 1, 3, 0, 3, 0, 0, 2, 6, 1, 0, 0, 0, 0, 1, 
7, 14, 23, 19, 10, 14, 29, 15, 7, 13, 16, 7, 9, 17, 6, 7, 16, 
6, 11, 13, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA), C_y1 = c(6, 9, 9, 2, 2, 4, 2, 2, 6, 
0, 6, 0, 1, 10, 3, 8, 14, 29, 15, 7, 17, 21, 24, 7, 32, 31, 31, 
21, 27, 29, 18, 27, 33, 23, 28, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA), D_y2 = c(6, 15, 
4, 1, 2, 4, 3, 1, 3, 0, 0, 0, 2, 2, 2, 5, 17, 21, 24, 7, 24, 
16, 28, 7, 28, 23, 25, 25, 24, 28, 33, 27, 31, 33, 21, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA), K_x1 = c(5, 1, 3, 1, 0, 6, 1, 2, 7, 1, 1, 0, 0, 2, 0, 8, 
24, 16, 28, 7, 24, 31, 31, 13, 32, 35, 32, 22, 29, 32, 32, 29, 
34, 32, 34, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA), L_x2 = c(5, 1, 3, 0, 3, 0, 0, 2, 6, 
1, 0, 0, 0, 0, 1, 7, 24, 31, 31, 13, 30, 30, 34, 12, 31, 27, 
23, 25, 33, 28, 31, 29, 30, 36, 24, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA), M_y1 = c(6, 
9, 9, 2, 2, 4, 2, 2, 6, 14, 23, 19, 10, 9, 10, 11, 14, 16, 8, 
24, 17, 8, 22, 14, 33, 28, 31, 14, 23, 19, 10, 9, 10, 11, 14, 
16, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA), N_y2 = c(3, 1, 0, 4, 0, 5, 6, 5, 17, 21, 24, 7, 
32, 31, 31, 21, NA, NA, NA, NA, 27, 29, 18, 27, NA, NA, 17, 21, 
24, 7, 32, 31, 31, 21, 27, 17, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA)), class = "data.frame", row.names = c(NA, 
-54L))

I need to separate it by subject and per conditions: one per column with the values of A, B, C, and D in one column and call it 'First' ; and K, L, M, N in another and call it 'Second'. More than this, the x, y, _1 and _2 present in these variables represent within-subject factors, that I also need to take into account into another two columns - 'Within1' for x and y; and 'Within2' for 1 and 2. And finally, with two columns 'Between1' and 'Between2' which are the between-subject factors.

I need it to look like this:

Subject First  SecondI  Within2  Within2 Between1 Between2
1         Ai     Ki        1      x          1        1 
1         Bi     Li        2      x          1        1 
1         Ci     Mi        1      y          1        1      
1         Di     Ni        2      y          1        1 
2         Ai     Ki        1      x          1        1 
2         Bi     Li        2      x          1        1 
2         Ci     Mi        1      y          1        1 
2         Di     Ni        2      y          1        1 
...

I have used the reshape function twice, once for grouping into one column, the A,B,C,D and separating the within-subject variables from it and I succeeded:



Subset_1 <-reshape(Subset_0, 
                   varying = c("A_x1", " B_x2", "C_y1", "D_y2"),
                   v.names = "First",
                   timevar = "Within1",
                   times =  c("A_x1", " B_x2", "C_y1", "D_y2"),
                   direction = "long")

# Next_Trial_Choice column
Subset_1$Within1[Subset_1$Within1== "A_x1"] <- "x"
Subset_1$Within1[Subset_1$Within1== "B_x2"] <- "x"
Subset_1$Within1[Subset_1$Within1== "C_y1"] <- "y"
Subset_1$Within1[Subset_1$Within1== "D_y2"] <- "y"

#cleaning the names - opponent column
Subset_1$Within2[Subset_1$Within2== "A_x1"] <- "1"
Subset_1$Within2[Subset_1$Within2== "B_x2"] <- "2"
Subset_1$Within2[Subset_1$Within2== "C_y1"] <- "1"
Subset_1$Within2[Subset_1$Within2== "D_y2"] <- "2"

The problem is that I need to do the same for another column ('Second') and I tried to use reshape again, as I did before, applied to the Subset1 this time. But it doesn't do what I need.

Is there a way to do this?

dataframe r
2 Answers

This looks like it gets your given example result:

# pipe
library(magrittr) 

# input data
dxyz <- structure(list(Subject = c(
  1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
  12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27,
  28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43,
  44, 45, 46, 47, 48, 49, 50, 51, 52, 54, 55
), BETWEEN1 = c(
  1,
  1, 1, 2, 2, 2, 2, 1, 1, 2, 2, 2, 2, 1, 2, 1, 2, 1, 1, 2, 2, 1,
  2, 1, 2, 1, 1, 2, 1, 1, 2, 1, 1, 2, 1, 2, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
), BETWEEN2 = c(
  1,
  1, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2,
  1, 1, 2, 2, 1, 1, 2, 2, 1, 1, 2, 2, 2, 2, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
), A_x1 = c(
  5,
  1, 3, 1, 0, 6, 1, 2, 7, 1, 1, 0, 0, 2, 0, 8, NA, NA, NA, NA,
  14, 23, 19, 10, 9, 10, 11, 14, 16, 8, 24, 17, 8, 22, 14, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
  NA, NA
), B_x2 = c(
  5, 1, 3, 0, 3, 0, 0, 2, 6, 1, 0, 0, 0, 0, 1,
  7, 14, 23, 19, 10, 14, 29, 15, 7, 13, 16, 7, 9, 17, 6, 7, 16,
  6, 11, 13, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA
), C_y1 = c(
  6, 9, 9, 2, 2, 4, 2, 2, 6,
  0, 6, 0, 1, 10, 3, 8, 14, 29, 15, 7, 17, 21, 24, 7, 32, 31, 31,
  21, 27, 29, 18, 27, 33, 23, 28, NA, NA, NA, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
), D_y2 = c(
  6, 15,
  4, 1, 2, 4, 3, 1, 3, 0, 0, 0, 2, 2, 2, 5, 17, 21, 24, 7, 24,
  16, 28, 7, 28, 23, 25, 25, 24, 28, 33, 27, 31, 33, 21, NA, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
  NA
), K_x1 = c(
  5, 1, 3, 1, 0, 6, 1, 2, 7, 1, 1, 0, 0, 2, 0, 8,
  24, 16, 28, 7, 24, 31, 31, 13, 32, 35, 32, 22, 29, 32, 32, 29,
  34, 32, 34, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA
), L_x2 = c(
  5, 1, 3, 0, 3, 0, 0, 2, 6,
  1, 0, 0, 0, 0, 1, 7, 24, 31, 31, 13, 30, 30, 34, 12, 31, 27,
  23, 25, 33, 28, 31, 29, 30, 36, 24, NA, NA, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
), M_y1 = c(
  6,
  9, 9, 2, 2, 4, 2, 2, 6, 14, 23, 19, 10, 9, 10, 11, 14, 16, 8,
  24, 17, 8, 22, 14, 33, 28, 31, 14, 23, 19, 10, 9, 10, 11, 14,
  16, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
  NA, NA, NA
), N_y2 = c(
  3, 1, 0, 4, 0, 5, 6, 5, 17, 21, 24, 7,
  32, 31, 31, 21, NA, NA, NA, NA, 27, 29, 18, 27, NA, NA, 17, 21,
  24, 7, 32, 31, 31, 21, 27, 17, NA, NA, NA, NA, NA, NA, NA, NA,
  NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
)), class = "data.frame", row.names = c(
  NA,
  -54L
))

# extract all abcd in long format with Within seperated
abcd <- dxyz %>% 
  tidyr::pivot_longer(-c(Subject, BETWEEN1,BETWEEN2)) %>% 
  tidyr::separate(col = name, sep = "_", into = c("First", "Within")) %>% 
  dplyr::filter(First %in% c("A", "B", "C", "D")) %>% 
  dplyr::mutate(
    Within21 = stringr::str_extract_all(Within, "[:digit:]") %>% unlist(),
    Within22 = stringr::str_extract_all(Within, "[:alpha:]") %>% unlist()
  ) %>% 
  dplyr::select(-Within)

# extract all klmn in long format with Within seperated
klmn <- dxyz %>% 
  tidyr::pivot_longer(-c(Subject, BETWEEN1,BETWEEN2)) %>% 
  tidyr::separate(col = name, sep = "_", into = c("Second", "Within")) %>% 
  dplyr::filter(Second %in% c("K", "L", "M", "N"))%>% 
  dplyr::mutate(
    Within21 = stringr::str_extract_all(Within, "[:digit:]") %>% unlist(),
    Within22 = stringr::str_extract_all(Within, "[:alpha:]") %>% unlist()
  ) %>% 
  dplyr::select(-Within)

# join both data sets together
abcd %>% 
  dplyr::left_join(
    klmn,
    by = c("Subject", "BETWEEN1", "BETWEEN2", "Within21", "Within22")
  ) %>% 
  dplyr::select(
    Subject, First, Second, Within21, Within22, BETWEEN1, BETWEEN2, value.x, value.y
  )

I seperated the reshaping into two pieces for for A, B, C, D and K, L, M, N and then joined the data together.

user_82826740

here is one option with pivot_longer. I know separated a bit too much, but it is just to remove confusion with names. You can adjust them according to your output.

library(tidyr)
df %>% pivot_longer(cols=c("A_x1", "B_x2", "C_y1", "D_y2"), names_to="first") %>%
  pivot_longer(cols=c("K_x1", "L_x2", "M_y1", "N_y2"), names_to="second",values_to = "value2") %>% 
  separate(first, into = c("first", "Within1"), sep = "_") %>% 
  separate(Within1,into = c("Within1", "Within1_2"), sep = "(?<=[A-Za-z])(?=[0-9])") %>% 
  separate(second, into = c("second", "Within2"), sep = "_") %>% 
  separate(Within2,into = c("Within2", "Within2_2"), sep = "(?<=[A-Za-z])(?=[0-9])") %>%  
  select(-c(value, value2)) %>% distinct()
user_103081910
Related