es_cdelta.RdNon-parametric effect size estimate used to quantify the degree to which two variable distributions overlap. Output is scaled to range between -1 to 1, with the arithmetic sign indicating if the comparator group lies to the right (higher) or left (lower) to a the reference.
es_cdelta(ref, comp)
| ref | num, reference group |
|---|---|
| comp | num, comparator group |
This function returns Cliff's delta effect size ranging from -1 to 1
Cliff, N (1993), Dominance statistics: Ordinal analyses to answer ordinal quetsions Psychological Bulletin, 114.(3), 494-509.https://doi.org/10.1037%2F0033-2909.114.3.494
Other NMR ++:
.hotellingsT2(),
cvanova(),
minmax(),
opls_perm(),
pca(),
plotscores(),
pqn(),
predict_opls(),
scRange()
torben.kimhofer@murdoch.edu.au
torben.kimhofer@murdoch.edu.au
# define two distrubutions a=rnorm(100, mean=0, sd=1) b=rnorm(100, mean=2, sd=1) # Cliff's delta es_cdelta(ref=a, comp=b)#> [1] 0.8494es_cdelta(ref=b, comp=a)#> [1] -0.8494