tag:blogger.com,1999:blog-21621108.post8374563111969005980..comments2023-03-25T14:52:12.967+01:00Comments on Shravan Vasishth's Slog (Statistics blog): A weird and unintended consequence of Barr et al's Keep It Maximal paperShravan Vasishthhttp://www.blogger.com/profile/13453158922142934436noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-21621108.post-17132052864226589002015-08-17T20:09:14.384+02:002015-08-17T20:09:14.384+02:00@Dani, Barr et al. say that you need random slopes...@Dani, Barr et al. say that you need random slopes for all fixed effects <i>about which you want to make inferences</i>. The last part is often forgotten. Specifically, this means that you don't need random slopes for covariates and this can be used to solve your problem: Assume a 2x2 design and a non-converging model with the following structure: y ~ a + b + (a+b|subj) + (a+b|item) Now, you can decompose this into two models:<br /><br />y ~ a + b + (a|subj) + (a|item)<br />y ~ a + b + (b|subj) + (b|item)<br /><br />Use the first model to make inferences about a (b is just a covariate there) and the second to make inferences about b (now a is the covariate). Since both models have a simpler random effects structure, there is a higher chance that they converge. Not sure if this approach is generally valid but it should at least work when a and b are orthogonal. Check the Barr et al. paper, I think they describe this approach somewhere in the discussion.Titus von der Malsburghttps://www.blogger.com/profile/14370107382857806010noreply@blogger.comtag:blogger.com,1999:blog-21621108.post-18112666598432070192015-03-23T14:15:16.945+01:002015-03-23T14:15:16.945+01:00Looking forward to it,Looking forward to it,Danihttps://www.blogger.com/profile/10128159415516788214noreply@blogger.comtag:blogger.com,1999:blog-21621108.post-86404630094530156942015-03-23T13:23:50.839+01:002015-03-23T13:23:50.839+01:00You should try to find the simplest model that is ...You should try to find the simplest model that is possible. Examples coming soon.Shravan Vasishthhttps://www.blogger.com/profile/13453158922142934436noreply@blogger.comtag:blogger.com,1999:blog-21621108.post-67277966531343048212015-03-18T13:07:56.449+01:002015-03-18T13:07:56.449+01:00If one is unable to fit a maximal model due to low...If one is unable to fit a maximal model due to low power, what should be done instead. For me, this is especially important in cases where neither IV1 or IV2 have a stronger case for being included over the other.Danihttps://www.blogger.com/profile/10128159415516788214noreply@blogger.comtag:blogger.com,1999:blog-21621108.post-78068032559374904282015-01-03T02:13:57.968+01:002015-01-03T02:13:57.968+01:00On the plus side, in the recent past many of those...On the plus side, in the recent past many of those authors probably would have just gone straight for the RM-ANOVA without trying a mixed model at all. So it's probably still progress.Anonymousnoreply@blogger.com