rankFD: An R Software Package for Nonparametric Analysis of General Factorial Designs

Abstract

Many experiments can be modeled by a factorial design which allows statistical analysis of main factors and their interactions. A plethora of parametric inference procedures have been developed, for instance based on normality and additivity of the effects. However, often, it is not reasonable to assume a parametric model, or even normality, and effects may not be expressed well in terms of location shifts. In these situations, the use of a fully nonparametric model may be advisable. Nevertheless, until very recently, the straightforward application of nonparametric methods in complex designs has been hampered by the lack of a comprehensive R package. This gap has now been closed by the novel R-package rankFD that implements current state of the art nonparametric ranking methods for the analysis of factorial designs. In this paper, we describe its use, along with detailed interpretations of the results.

Performance of Aubergine Rootstocks against Verticillium dahliae Isolates in Southeastern Spain

“…Symptoms were also analysed by using rank-based non-parametric methods for the analysis of ordinal data in factorial designs [35]. From the midranks of the observations, the non-parametric relative effect of treatment was estimated, and its confidence intervals and the significance of the factors Solanum rootstock and V. dahliae isolate and their interaction were tested with the ANOVA-type statistic by using the rankFD function of the rankFD software package [36,37].…”

Recognition and perception of emotions in juvenile myoclonic epilepsy

“…28 Univariate comparisons between groups were done via a nonparametric t-test (Brunner-Munzel test) using the R package rankFD. 27 Kendall's tau was computed and tested for significance via the package Kendall 26 ; confidence intervals were obtained via bootstrap. 29 The main effect measure was the nonparametric relative treatment effect (RTE).…”

Standard vs. carbone dioxide adapted kidney replacement therapy in hypercapnic ARDS patients: a randomized controlled pilot trial (BigBIC)

“…To compare metric outcomes between control and intervention group, the Brunner-Munzel test was used due to non-normally distributed values with heterogeneous variances. The test was performed using the function rank.two.samples() from the R package rankFD [ 13 ]. The effect measure of the Brunner-Munzel test is the relative effect p with the null hypothesis p = 1/2.…”