RESI is an R package designed to implement the Robust Effect Size
Index (RESI, denoted as S) described in Vandekar, Rao, & Blume
(2020). The RESI is a versatile effect size measure that can be easily
computed and added to common reports (such as summary and ANOVA tables).
This package currently supports lm
, glm
,
nls
, survreg
, coxph
,
hurdle
, zeroinfl
, gee
,
geeglm
, lme
, and lmerMod
models.
Nonparametric bootstrapping is used to compute confidence intervals,
although the interval performance has not yet been evaluated for the
longitudinal models. A Bayesian bootstrap is also available for
lm
and nls
models. In addition to the main
resi
function, the package also includes a
point-estimate-only function (resi_pe
), conversions from S
to other common effect size measures and vice versa, print methods, plot
methods, summary methods, and Anova/anova methods. A more detailed
vignette is being written.
Vandekar S, Tao R, Blume J. A Robust Effect Size Index. Psychometrika. 2020 Mar;85(1):232-246. doi: 10.1007/s11336-020-09698-2.
Kang, K., Armstrong, K., Avery, S., McHugo, M., Heckers, S., & Vandekar, S. (2021). Accurate confidence interval estimation for non-centrality parameters and effect size indices. arXiv preprint arXiv:2111.05966.