Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Quick Links

Categories

Supported by
Bestel Een wereld vol denkers door Sebastiaan Mathôt via bol.com

JASP Results not comparable to MATLAB or R

alessmanoloalessmanolo
in JASP & BayesFactor

Hi there,

so I am running the same Data through a GLMM in JASP, MATLAB & R to get an overview/confidence in my results. Its a simple Generalized Linear Mixed effect model on reaction times being influenced by the stimulus type and the movement type that was to be performed.

While MATLAB and R have congruent results JASP shows some oddities. As to be seen in the attached images. Especially odd is the difference in DF, where MATLAB consistently choose the number of trials (7063) and JASP chooses the Number of participants (~23) for the intercept. This is very odd and does not react to any changes in the model.

If you have any idea why this is the case, please let me know!

JASP_1.png
JASP_2.png
MATLAB.png


Comments

  • MSBMSB

    Two things are going on:


    1. JASP is using either Satterthwaite or Kenward-Rogers estimates of degrees of freedom, while MATLAB and lme4 use residual dfs or no dfs (in R you can use lmerTest instead to also get Satterthwaite / Kenward-Rogers dfs).
    2. JASP is using effect coding for the dummy variables while by default R and MATLAB use treatment coding for dummy variables.
Sign In or Register to comment.
Sebastiaan Mathôt » v2.0.18 » Copyright 2011 — 2019 » Vanilla flavored » License