Dear Yannis,
A comparison of the two approaches is available in this book chapter
from Rik Henson:
http://www.mrc-cbu.cam.ac.uk/wp-content/uploads/2015/03/Henson_EN_15_ANOVA.pdf
A key advantage of the partitioned error model here, with one sample
t-tests, is that you completely eschew the non-sphericity problem.
Best regards,
Guillaume.
On 06/06/18 09:15, Yannis Paloyelis wrote:
> Dear Guillame,
>
> With respect to this 2x2 repeated measures design, I wonder if you could
> explain: is there any advantage (and what is it) of using a partitioned
> errors model with a series of appropriately constructed contrasts and
> one sample t-tests, versus a flexible factorial (and pool error variance)?
>
> Many thanks for your advice
>
> Yannis
>
>
>
>
> On 7 May 2018 at 15:56, Guillaume Flandin <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> Dear Sophie,
>
> You have two within-subjects factors with two levels each here so I
> would recommend to compute a contrast image for each main effect and
> interaction for each subject and enter these in a one-sample t-test at
> the second level, ie at first level, compute:
> main effect of drug: [1 1 -1 -1]
> main effect of cardiac: [1 -1 1 -1]
> drug x cardiac interaction: [1 -1 -1 1]
>
> Best regards,
> Guillaume.
>
>
> On 03/05/18 09:31, Sophie Betka wrote:
> > Dear Spm users,
> >
> >
> >
> > Could you let me know if what I did is correct? Thanks a lot
> >
> >
> >
> > So I am using a 2*2 flexible design (within-subjects design with
> > Subjects, Drug:2 Levels ; CardiacTiming:2 Levels).
> >
> >
> >
> >
> >
> > 1) I attached a screen shot of my batch. The variance of Drug and
> > Cardiac should be unequal right?
> >
> > I am interested in main effect of Drug, main effect of
> > Cardiac and the interaction.
> >
> >
> >
> > 2) I attached a screen shot of the design matrix.
> >
> > Following the Glascher’s paper, I created contrasts.
> >
> >
> >
> > Could confirm/correct my contrasts:
> >
> >
> >
> >
> >
> >
> >
> > Drug1
> >
> >
> >
> > Durg2
> >
> >
> >
> > Card1
> >
> >
> >
> > Card2
> >
> >
> >
> > D1C1
> >
> >
> >
> > D1C2
> >
> >
> >
> > D2C1
> >
> >
> >
> > D2C2
> >
> > Main effect of Drug
> >
> >
> >
> > 1
> >
> >
> >
> > -1
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 0.5
> >
> >
> >
> > 0.5
> >
> >
> >
> > -0.5
> >
> >
> >
> > -0.5
> >
> > Main effect of Cardiac
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 1
> >
> >
> >
> > -1
> >
> >
> >
> > 0.5
> >
> >
> >
> > -0.5
> >
> >
> >
> > 0.5
> >
> >
> >
> > -0.5
> >
> > Interaction Drug*Cardiac
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 0.5
> >
> >
> >
> > -0.5
> >
> >
> >
> > -0.5
> >
> >
> >
> > 0.5
> >
> > Test single repressor Drug1
> >
> >
> >
> > 1
> >
> >
> >
> > 0
> >
> >
> >
> > 0.5
> >
> >
> >
> > 0.5
> >
> >
> >
> > 0.5
> >
> >
> >
> > 0.5
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> > Test single repressor D2C2
> >
> >
> >
> > 0
> >
> >
> >
> > 1
> >
> >
> >
> > 0
> >
> >
> >
> > 1
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 0
> >
> >
> >
> > 1
> >
> >
> >
> > The main effect of cardiac gives me a massive cluster; this a bit odd
> > (enclosed).
> >
> >
> >
> >
> >
> > Thank you for your help,
> >
> > S.
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > *Sophie BETKA*
> > *Ph.D Student*
> > *Neuropsychologist*
> >
> > *Brighton and Sussex Medical School*
> >
> > Neurosciences Department - CritchleyLab
> >
> > /Trafford Centre, 202 (Brighton, UK)/
> >
> > University of Sussex, Falmer, BN1 9RR
> > +44 (0)1273 873132
> >
> > *SCALab - Cognitive and Affective Sciences Laboratory*
> > /DEEP Team - Emotional Dynamics and Pathology (Lille, FR)/
> > Faculté de Medecine, Pôle Recherche
> > 1, Place de Verdun, 59045 Lille cedex
> >
> > +33 (0)3.20.62.68.13
> >
> > _Twitter @SophieBetka <https://twitter.com/SophieBetka
> <https://twitter.com/SophieBetka>>___
> >
> > _ResearchGate <https://www.researchgate.net/profile/Sophie_Betka
> <https://www.researchgate.net/profile/Sophie_Betka>>___
> >
> > _Linkedin <https://www.linkedin.com/in/sophie-betka-54b149155/
> <https://www.linkedin.com/in/sophie-betka-54b149155/>>___
> >
> >
> >
> >
> >
> >
> > ___________________________________________________________
> > This email has been scanned by MessageLabs' Email Security System
> > on behalf of the Brighton & Sussex Medical School. For more
> information see:
> > https://staff.brighton.ac.uk/is/computing/Pages/Email/spam.aspx
> <https://staff.brighton.ac.uk/is/computing/Pages/Email/spam.aspx>
>
> --
> Guillaume Flandin, PhD
> Wellcome Centre for Human Neuroimaging
> University College London
> 12 Queen Square
> London WC1N 3BG
>
>
--
Guillaume Flandin, PhD
Wellcome Centre for Human Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
########################################################################
To unsubscribe from the SPM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=SPM&A=1
|