Cohen's kappa for a two-by-two table, with a large-sample interval

Uses SE = sqrt(po (1 - po) / (n (1 - pe)^2)). With a small denominator the interval is wide, which is the honest answer rather than a defect.

Usage

kappa_ci(tp, fn, fp, tn, conf_level = 0.95)

Arguments

tp, fn, fp, tn

Counts of true positives, false negatives, false positives and true negatives. Vectorised and recycled together.

conf_level

Confidence level.

Value

A three-column matrix with columns est, lo and hi.

Examples

kappa_ci(tp = 80, fn = 20, fp = 10, tn = 90)
#>      est        lo        hi
#> [1,] 0.7 0.6010267 0.7989733