External Email - Use Caution

Hello Jessica,

please see my inline reponses below.

Best regards,

Kersten

On Mo, 2019-12-02 at 02:48 +0000, Hua, Jessica wrote:

External Email - Use Caution         Hi FreeSurfer Experts,

I used the longitudinal pipeline and analyzed my data using the univariate linear mixed effects models.  I had a couple of questions:

1. The model outputted the F and p values.  How do I calculate the

effect size?

There are some standardized effect sizes for LME models described in the literature, but none are, as far as I know, included in the lme toolbox, so one would still need to implement them.

In my opinion, it's therefore easiest to express the magnitude of changes / differences in terms of unstandardized effect sizes, i.e. the beta coefficients estimated by the model.

2. How do I plot my analyses?  I would like to create a plot for

change in CC volume with eBAC. Below is my model:

Yij = ß1 + ß2(timeij) + ß3(eBACi) + ß4*estimated eBACi*timeij + ß5(genderi) ß6(intracranial volumei) + b1i + b2i(timeij) + eij.

That really depends on how your plot should look like, and may require some custom Matlab programming. For 'spaghetti'-style plots (i.e. a bunch of individual slopes), have you explored the 'lme_timePlot' function?

Thank you for your help!

Jessica

---

Jessica Hua, M.A. Cognitive and Emotional Control Lab Doctoral Candidate Clinical Psychology  University of Missouri _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer