Wald test that a coefficient is equal across two or more fits, or equal to
value for a single fit. With more than two fits it tests that all are
equal (G - 1 degrees of freedom).
Fits are independent samples by default, so their variances add. Use
paired = TRUE for fits on the same subjects (two estimators, two
attributes, two rater groups, ...): the dependence then comes from the
per-subject influence rows, matched by the row names of the rating data, or
by row number when there are none.
Arguments
- ...
Fits from
agreement()oragreement_counts(). Names label the output.- coef
Coefficient to compare. Defaults to the first one (
"Conger"foragreement(),"Fleiss"foragreement_counts()).- paired
FALSE(default) for independent samples,TRUEfor fits on the same subjects.- value
Null value for the coefficient (one fit) or the difference (two fits). Default
0.
Examples
# Two studies with different subjects.
agreement_test(gwet = agreement(dat.gwet2014),
klein = agreement(dat.klein2018))
#> Warning: rater pair(s) rater4-rater5 co-observed by only one subject; the corresponding pairwise covariance is degenerate and the standard error unreliable.
#>
#> Independent-sample Wald test of equal Conger across 2 fits
#>
#> data: gwet, klein
#> X-squared = 0.0019646, df = 1, p-value = 0.9646
#> alternative hypothesis: true difference in Conger is not equal to 0
#> sample estimates:
#> gwet klein
#> 0.4233051 0.4301049
#>
# Two rater pairs on the same 50 subjects.
agreement_test(AB = agreement(dat.zapf2016[, 1:2], estimator = "pairwise"),
CD = agreement(dat.zapf2016[, 3:4], estimator = "pairwise"),
paired = TRUE)
#>
#> Paired Wald test of equal Conger across 2 fits
#>
#> data: AB, CD
#> X-squared = 2.2432, df = 1, p-value = 0.1342
#> alternative hypothesis: true difference in Conger is not equal to 0
#> sample estimates:
#> AB CD
#> 0.9309091 0.8466753
#>
