Stats Apps Tutorials: 23. How to run Linear Mixed Effects Models in SPSS, JASP, and R

Links to video sections and data files are in the description below. In this tutorial video we go through the steps to perform Linear Mixed Effects (LME) analysis using SPSS, JASP and R / RStudio.

Data files used in this video:

Sections:
00:00​ Introduction
00:29 Running LME analysis in SPSS
03:35 Interpreting LME output in SPSS
06:07 Running LME analysis in JASP
10:11 Setting up for LME analysis in R
13:37 Running LME analysis in R
16:22 Interpreting LME output from R
17:35 Plotting the data and looking at pairwise contrasts
21:34 LME diagnostics in R

[charlesgervais]

Only video I've found that clearly explains how to choose random and fixed effects and how to navigate SPSS properly. Thank you so much

[lureennaser]

Wow! what a clear easy to follow tutorial for beginners like me. Thank you so much. You are a life savior.

[rongruihuo]

TNice tutorials is absolutly the best video of the world you expaining skills are good and it was a honor to see tNice tutorials vid well done

I found the R usage for mixed effects in this video. Thank you Tony.

[MikkoHaavisto]

Thank you for being so clear and consise. Easy to follow along.

[rekabuzassy]

Great video! Do you know perhaps how can I plot my significant interaction in SPS LMM?

[Bbb-uuy]

Hi Thank you for the video! What if attitude is a continuous variable?

[theflamekid]

Hi, I have a question. You now use 3 nominal variables in the model, but what if you use a continous scale as variable? How do I put that in in SPSS? Because right now, it does not look right it...

[osamamahmoudm.alomari]

Thank you for this easy to follow tutorial, but I have a question. You mentioned that "SPSS does not allow us to do post hoc tests for interactions", so how can we do post hoc tests for significant interactions when we have a three-level independent factor???

[MaryanneNduka]

nice video, please do you have a SAMPLE document on how to report LMM.

[nanli]

Great video and clear explanations! I have a question. In 20:02, the statistics in lmer in R showed that P values for SexMale and Attitudepolite were 0.00376 and 0.00311, respectively. However, the anova output for the P values of the two factors were 0.006808 (sex) and 0.003402 (attitude). So, what is the difference between the P values from lmer and anova?

[stinkbomb]

how would i do final visualization of a mixed effects model? and are residuals and/or predicted values typically visualized too?..

[niziolek]

Very good tutorial.
I have a problem with running LMM for estimated data. I have some missing data in my longitudinal experimental study. I've estimated them with SPSS, but then in LMM results referring to Type III Tests of Fixed Effects are presented only for each imputation separately and not for combined data. Do you know any option to get results based on combined data from all (20) imputations?

[barbarasimek]

hi. i have a question. How can we interpret the outputs in jasp?

[MomoSimone]

Hi there! Great video! Would you have any information on how I can find the standard deviations for my estimated marginal means for the post-intervention results (i.e., time point 2)? SPSS does not give me these descriptives, but I need to report them in my paper.

[melaniebenard]

Hi there! thank you for your video; I have one question, at 4.46 interpreting the output in SPSS; the F statistic is significant for attitude (p = 0.003) but the t statistic in estimates of fixed effects is not (p = 0.221). How is this possible? I have the same problem with my own data..

[MrEgwEdw]

Hello, quick question, in my fixed effects analysis, where fixed factors were time and industry group, both of these categories had p>.05. When writing the equation, do I include them?

[staceyreyes]

What version of R is this? I get a warning that says package "lmer" is not available for R version 3.6.3

[mleoconnor]

Hi! This video was super helpful. I'm wondering if you would be able to do a similar one where the items are correlated?

[monaadnan]

How can we test for normality in Linear Mixed Model, in SPSS .. and do we report the p-values of the type 3 tests of fixed effects or the estimates of fixed effects.. Thanks in advance!