I’m trying to calculate multiple ICCs from a data frame created in R. The data frame contains over 2500 observations from 122 different persons. I need an ICC calculated for each person.

Creating a separate data frame for one person, I am able to calculate the ICC using the package irr in R:

data_edue_icc_212003 <- subset(data_edue, vp_id == 212003 , 
                        select = c(stress_sad : stress_tense))

icc(data_edue_icc_212003,
    model = "twoway",
    type = "agreement",
    unit = "single")

These are the results:

Single Score Intraclass Correlation

   Model: twoway 
   Type : agreement 

   Subjects = 19 
     Raters = 9 
   ICC(A,1) = 0.214

 F-Test, H0: r0 = 0 ; H1: r0 > 0 
 F(18,89.3) = 4.37 , p = 1.35e-06 

 95%-Confidence Interval for ICC Population Values:
  0.095 < ICC < 0.42

Instead of having to subset the data frame 122 times to isolate the data for one person, I want to apply ICC by row, grouping the data frame by ID.
I guess there is a quicker way to calculate ICCs all at once than calculating 122 ICCs on separate data frames. What I’m looking for is an output that could look like this:

Test    ICC
 ID1   (icc value)
 ID2   (icc value)
...