Computes the intraclass correlation coefficient ICC(A,1) with its confidence interval and the overall concordance correlation coefficient (OCC) among raters for each condition, as in Table 3 of Franceschi et al. (2020).
Arguments
- x
An object returned by
sad_data().- conf
Confidence level for the ICC interval.
Value
A tibble with one row per condition and columns ICC,
ICC_lower, ICC_upper, OCC, n_raters and n_images.
Details
ICC(A,1) is the two-way random effects, absolute agreement, single rater
coefficient of McGraw and Wong (1996). OCC is the overall concordance
correlation coefficient of Barnhart et al. (2002), computed as in
epiR::epi.occc(). Only raters that assessed
all images of a condition are used.
References
McGraw, K. O., and Wong, S. P. (1996). Forming inferences about some intraclass correlation coefficients. Psychological Methods 1:30-46.
Barnhart, H. X., Haber, M., and Song, J. (2002). Overall concordance correlation coefficient for evaluating agreement among multiple observers. Biometrics 58:1020-1027.
Examples
x <- sad_data(SADRustSoybean, wide = c(UN = "unaided", SAD = "aided"), group = "sad")
sad_reliability(x)
#> # A tibble: 4 × 7
#> condition ICC ICC_lower ICC_upper OCC n_raters n_images
#> <fct> <dbl> <dbl> <dbl> <dbl> <int> <int>
#> 1 UN_Old 0.749 0.650 0.833 0.747 20 50
#> 2 UN_New 0.739 0.644 0.823 0.736 20 50
#> 3 SAD_Old 0.762 0.673 0.840 0.760 20 50
#> 4 SAD_New 0.942 0.918 0.962 0.941 20 50