RcmdrPlugin.aRnova: a Tutorial (in progress…)

Jessica Mange and Arnaud Travert


1. Introduction

RcmdrPlugin.aRnova is a R Commander add-on for repeated-measures and mixed-design (‘split-plot’) analysis of variance (ANOVA). In essence it adds a new menu entry for repeated measures that allows to deal with up to three within-subject factors and optionally with one or several between-subject factors. Besides, it also provides supplementary options and outputs to the existing One Way ANOVA and Multi Way ANOVA entries, such as the choice of type of sum of squares (II or III), the calculation of effect sizes, or multiway ANOVA post-hoc analysis.

This tutorial (not finished yet…) aims to illustrate typical use of this plugin by replicating ANOVAs described in particular in John Fox’ R and S-plus Companion to Applied Regession (SAGE Publications Inc. 2002). While all steps are illustrated in the following, the user is expected to have minimal proficiency in using Rcmdr.

2. Installation and loading of RcmdrPlugin.aRnova

The aRnova plugin can be downloaded and installed using:

> install.packages("RcmdrPlugin.aRnova")

To load the package the fastest way is to use the following R command line which causes the package to load Rcmdr simultaneously with the aRnova plugin:

> library("RcmdrPlugin.aRnova")

Note that if aRnova has just been installed, restart a new R session before launching this command (else the following error may occur: Error : .onAttach failed in attachNamespace() for [....]).

Alternatively, after loading Rcmdr without the aRnova plugin using:

> library("Rcmdr")

The aRnova package can then be loaded using the Tools > Load Packages... menu entry in Rcmdr. The user is then asked to restart Rcmdr so that the Plug-in is available.

It can be verified that aRnova has been effectively loaded by checking that the supplementary entry Repeated measures ANOVA... has been added in the Statistics > Means Rcmr menu.

3. One-way ANOVA

To be completed….

4. Multi Way ANOVA

The multiway ANOVA is illustrated here using the data form Moore and Krupat (1971) on conformity. The dataset can be found in the car package (Moore.rda) and in the present package for convenience. The treatment below reproduces the treatmant of section 4.3 of John Fox R and S-plus Companion to Applied Regession (SAGE Publications, 2002).

Figure 3 below illustrates how to (1) load and (2) view the Moore dataset which first entries are shown in (3). In this experiment, the conformity (DV, 1st columbns) of each subject interacting with a partner of either ‘low’ or ‘high’ status (partner.status, 1st IV, 2nd column) and’low’, ‘medium’ or ‘high’ authoritarism (fcategory, 2nd IV, 3rd column).