Visualizing three-way interaction
Dear all,
thanks for the many helpful suggestions. I decided to go with Ulf's
approach to extract the values from the effects plot; it is very easy and
the results should satisfy these 2-dimensional reviewers (I hope!). Just
wanted to correct a typo in Tom's code:
ne.effect$neuroticismF <- factor(ne.effect$neuroticism,
labels = paste0("extraversion: ",
unique(ne.effect$neuroticism)))
I guess the pasted label should read "Neuroticism: " (correct me if I'm
wrong). Thanks again for all the help and happy Easter!
Klemens
On 13 April 2017 at 11:03, Houslay, Tom <T.Houslay at exeter.ac.uk> wrote:
Hi Klemens, You probably have all the answers you need now, but just in case they are at all useful then I have a little series of posts on my website about visualising 3-way interactions using various methods: 2 continuous, 1 categorical: https://tomhouslay.com/2015/06/02/understanding-3-way- interactions-between-continuous-and-categorical- variables-part-ii-2-cats-1-con/ 1 continuous, 2 categorical: https://tomhouslay.com/2014/09/06/understanding-3-way- interactions-between-continuous-and-categorical-variables-small-multiples/ 3 continuous: https://tomhouslay.com/2014/03/21/understanding-3-way- interactions-between-continuous-variables/ Some of these are a little old and don't take advantage of better functions in R (that didn't exist or I was unaware of at the time!), but might be helpful for ideas on different ways of plotting things. These are generally using standard regression models, but predict/broom etc can be used to average over random effects (eg using 're.form = NA' in the predict.merMod function). Good luck with your plots! Cheers Tom ------------------------------ *From:* R-sig-mixed-models <r-sig-mixed-models-bounces at r-project.org> on behalf of r-sig-mixed-models-request at r-project.org < r-sig-mixed-models-request at r-project.org> *Sent:* 13 April 2017 09:39 *To:* r-sig-mixed-models at r-project.org *Subject:* R-sig-mixed-models Digest, Vol 124, Issue 12 Message: 1 Date: Wed, 12 Apr 2017 11:41:00 +0200 From: Klemens Kn?ferle <knoeferle at gmail.com> To: r-sig-mixed-models at r-project.org Subject: [R-sig-ME] LMER: Visualizing three-way interaction Message-ID: <CAANFXSJgSG7oUBDhY-+dDuvrFUWfBNm=TP2BZ20QCbYOj5KPL w at mail.gmail.com> Content-Type: text/plain; charset="UTF-8" Hi all, I'm trying to visualize a three-way interaction from a rather complex linear mixed model in R (lmer function from the lme4 package; the model has a complex random-effects structure). The interaction consists of two continuous variables and one categorical variable (two experimental conditions). So far, I have graphed the interaction via two 3D-surface plots using visreg2d from the visreg package. But my reviewers found these plots confusing and asked for a different illustration, such as conditional coefficient plots (i.e., plots of the strength of coefficient 1 as coefficient 2 increases). I've tried to find a package that allows me to create these kind of plots, but failed. The existing packages only allow coefficient plots for two-way interactions (for instance the interplot package; https://cran.r-project.org/web/packages/interplot/ vignettes/interplot-vignette.html). That means I only get a conditional coefficient plot of the two-way interaction, collapsed across both levels of the categorical variable. Is there a package for my case? If not, I probably have to manually extract fitted values from my model (e.g., using broom) and somehow plot them in ggplot2. But I don't really know how to do this, whether or not to take into account random effects (and how), etc. Any ideas would be much appreciated... Klemens Kn?ferle, Ph.D. Associate Professor - Department of Marketing BI Norwegian Business School Visiting address: Nydalsveien 37, 0484 Oslo [[alternative HTML version deleted]]