Matching and analysing data using the purr package in R

Question

I want to analyse my data using the purr package and psych package. Here is part of my data:

df <- tribble(
    ~temp1, ~temp2, ~temp3, ~temp4, ~temp5, ~temp6, ~temp7, ~temp8,
    75, 88, 85, 71, 98, 76, 71, 57,
    80, 51, 84, 72, 59, 81, 70, 64,
    54, 65, 90, 66, 93, 88, 77, 59,
    59, 87, 94, 75, 74, 53, 56, 87,
    52, 55, 64, 77, 50, 64, 83, 87,
)

Here, I want to match pay1 with pay2, pay3 with pay4 and pay5 with pay6 only. I could do it using the purr package for some methods, for example, correlation, I could use the following codes:

df %>% 
  split.default(rep_len(1:2, ncol(.))) %>% 
  pmap_dbl(~ cor( .x,.y))

But It does not work for ICC in the psych package

df %>% 
  split.default(rep_len(1:2, ncol(.))) %>% 
  pmap_dbl(~ ICC( .x,.y))

Can we do ICC using the psych package, basic r or other packages?

Answer 1

We can use pmap here since ICC doesn't return numeric values but an output of class "psych" "ICC" ..

library(purrr)
library(psych)

df %>%  split.default(rep_len(1:2, ncol(.))) %>% pmap(~ICC(cbind(..1, ..2)))

#$temp1
#Call: ICC(x = cbind(..1, ..2))

#Intraclass correlation coefficients 
#                         type  ICC   F df1 df2    p lower bound upper bound
#Single_raters_absolute   ICC1 0.12 1.3   4   5 0.39       -0.61        0.78
#Single_random_raters     ICC2 0.12 1.3   4   4 0.41       -0.65        0.78
#Single_fixed_raters      ICC3 0.12 1.3   4   4 0.41       -0.67        0.78
#Average_raters_absolute ICC1k 0.21 1.3   4   5 0.39       -3.11        0.87
#Average_random_raters   ICC2k 0.21 1.3   4   4 0.41       -3.68        0.88
#Average_fixed_raters    ICC3k 0.21 1.3   4   4 0.41       -4.06        0.88

# Number of subjects = 5     Number of Judges =  2
#$temp3
#Call: ICC(x = cbind(..1, ..2))

#Intraclass correlation coefficients 
#                         type   ICC    F df1 df2    p lower bound upper bound
#Single_raters_absolute   ICC1 -0.24 0.61   4   5 0.67       -0.79        0.59
#...
#...