Shrout and Fleiss (1979) Case 1: (MSR - MSW)/MSR, the
reliability of the MEAN of k ratings when each target is rated by
a DIFFERENT randomly chosen set of raters. Rater identity is not
crossed with target, so no rater effect is estimable and MSW pools
rater with error – which is why ICC(1,*) is the SMALLEST of the
three cases on the same data. If the same raters rated everyone,
this is the wrong coefficient.
Usage
morie_psy_icc1k(y, cluster, rater = NULL)
Arguments
- y
ratings.
- cluster
target identifier.
- rater
accepted for signature parity; Case 1 does not model
rater identity and the output records that it was ignored.
Value
list: value (ICC(1,k)), icc_single, k, n, MSR, MSW, case,
design_assumption, method.
References
Shrout and Fleiss (1979), Psychological Bulletin
86:420-428, Case 1 and Table 4.
Examples
morie_psy_icc1k(c(9, 2, 5, 8, 6, 1, 3, 2), rep(1:2, each = 4))$value
#> [1] 0.5925926
