Dear Doug
this is a snapshot of the website (https://surfer.nmr.mgh.harvard.edu/fswiki/LinearMixedEffectsModels) where the categorical variable "group" is coded as a number (see point 4)
****************************************************************************************** Example data analyses
Here are two example analyses using ADNI data processed with Freesurfer.
a)Univariate
In this case the interest was in determining the differences in mean hippocampal volume change over time among four groups of individuals: healthy controls (HC), stable mild cognitive impairment patients (sMCI), MCI patients who converted to Alzheimer's disease (cMCI) during the follow-up period and patients with Alzheimer's disease (AD) at baseline. A Qdec table file, qdec.table.dat, was used to process in Freesurfer all time point scans from all subjects included in the study. It contains the following columns:
1) Freesurfer's fsid (fsid)
2) subject ID (fsid-base)
3) time from baseline (time)
4) group (HC=1, sMCI=2, cMCI=3, AD=4)
5) Apoe4 (E4:carriers=1, non-carriers=0)
6) Gender (female=1, male=0)
7) Age at baseline (BslAge)
8) Education (in years).
***************************************************************************************************
In the case reported on the website, there were 4 groups, and groups were encoded with the following numbers: HC=1, sMCI=2, cMCI=3, AD=4.
In my analysis, I have just two groups (HC and Patients). If I put HC=1, Patients=2 (following the example of the website), I obtain some results from the LME model. However, if I center around 0 the "group" covariate, (i.e., HC=1, Patients=-1), I obtain different results from LME estimation. Therefore, I was wondering whether it is more correct to de-mean the "group" covariate or not.
Thank you for any suggestion Kind regards Paola
-----Messaggio originale----- Da: freesurfer-bounces@nmr.mgh.harvard.edu [mailto:freesurfer-bounces@nmr.mgh.harvard.edu] Per conto di Douglas N Greve Inviato: 29 September 2016 18:37 A: freesurfer@nmr.mgh.harvard.edu Oggetto: Re: [Freesurfer] Demeaning of "group" covariate in LME models
sorry, not sure I understand. Are you coding a catagorical variable as a continuous covariate? What tutorial are you referring to?
On 09/29/2016 10:39 AM, Valsasina Paola wrote:
Dear all,
I am trying to perform a longitudinal statistic by using the LME mass univariate toolbox. I have two groups of subjects (each acquired 3 or 4 times over the follow-up), and I want to test whether atrophy progresses differently over time in these two groups.
I followed the tutorial on the Website, and I used the following design matrix to test the time x group interaction: X=[ones(length (M),1) M M(:,1).*M(:,2)]
I first tried to code the categorical variable “group” assigning values 1 and 2 for subjects belonging to group 1 and 2, respectively (as it is suggested also in the tutorial). Then, I tried to re-run the same model by using mean-centered values for “group” (i.e., assigning -1 to subjects of group 1, and 1 to subjects of group 2).
When I run the two statistical models, I obtain different results. Therefore, I would like to ask which is the correct way to model the group covariate? Should it be zero-centered or not? How de-meaning is affecting the estimate?
Thank you so much for any suggestion
Kind regards
Paola
/Rispetta l’ambiente: non stampare questa mail se non è necessario. Respect the environment: if it's not necessary, don't print this mail./
IL TUO 5XMILLE AL SAN RAFFAELE DI MILANODevolvi il tuo 5 per mille all’Ospedale San Raffaele: perché al centro della Ricerca ci sei TU. CODICE FISCALE: 07636600962, nel riquadro RICERCA SANITARIA. Non c’è cura, senza ricerca. Non c’è ricerca, senza il tuo 5xmille. Scopri come su http://www.5xmille.org
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Rispetta l’ambiente: non stampare questa mail se non è necessario. Respect the environment: if it's not necessary, don't print this mail.
IL TUO 5XMILLE AL SAN RAFFAELE DI MILANO
Devolvi il tuo 5 per mille all’Ospedale San Raffaele: perché al centro della Ricerca ci sei TU. CODICE FISCALE: 07636600962, nel riquadro RICERCA SANITARIA. Non c’è cura, senza ricerca. Non c’è ricerca, senza il tuo 5xmille. Scopri come su http://www.5xmille.org