[Eeglablist] Including interaction with condition with two datasets per participant

[Dr Cyril, Pernet]

Hi Agata

alternatively you can set the two sessions in STUDY. Each session with 
language and valence. STUDY/LIMO will automatically compute the 
difference between sessions, ie you end up with contrasts language 
difference and valence difference. This allows setting up and ANOVA 2x2 
on those difference (of course each effect now reflects the 
difference/interaction with relationship which is like having an ANOVA 
with only interaction terms). The main differences with merging are (1) 
does not assume the same variance between sessions (2) but does not 
allow testing main effects across sessions (although you can do the 2x2 
in each session to check consistency of results).

Cyril


On 07/03/2025 21.08, Cedric Cannard via eeglablist wrote:
> Hi Agata,
>
> If I understand correctly, you need to merge the two datasets for each subject, ensuring that markers mark the conditions differently. Then, in study design, you can select the conditions of interest, and create different ones for different statistical analyses in LIMO.
>
> Hope this helps,
>
>
> Cedric
>
>
>
> On Friday, March 7th, 2025 at 4:21 AM, Agata Dymarska via eeglablist <eeglablist at sccn.ucsd.edu> wrote:
>
>> Hi,
>>
>> We are trying to analyse data using LiMo and compare 3 conditions, all within-participants (language, valence, relationship). One of the conditions (relationship status) was manipulated between sessions, such that we have two datasets for each participant. We are having trouble including this condition in the analysis.
>> In STUDY the subject names are the same for both sessions, but condition is coded as different in each session. However, at the stage of creating a design, the design can include two conditions as well as 4 variables of different types, but there is no way to cross them into a 2x2x2 design. This becomes a problem in level 2 analysis, where repeated measures ANOVA can only be performed on the type variables (Language and Valence), without the inclusion of the 3rd condition (Relationship).
>>
>> What would be the best way to code a condition that is different (and counterbalanced) in different sessions with the same participants?
>>
>> Thank you
>> Agata Dymarska
>> _______________________________________________
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu or visit https://sccn.ucsd.edu/mailman/listinfo/eeglablist .
> _______________________________________________
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu or visit https://sccn.ucsd.edu/mailman/listinfo/eeglablist .

-- 
Dr Cyril Pernet, PhD, OHBM fellow, SSI fellow
Neurobiology Research Unit,
Building 8057, Blegdamsvej 9
Copenhagen University Hospital, Rigshospitalet
DK-2100 Copenhagen, Denmark

wamcyril at gmail.com
https://urldefense.com/v3/__https://zcal.co/cpernet__;!!Mih3wA!CiwC6zfeIyr6DIp0qqUZNapI4qIaSCYUAhPtwuWfaKJ_wGfhuYyZrGljygq6qYUY__bSvW_nN3mL3SbPZJvXHQ$ 
https://urldefense.com/v3/__https://cpernet.github.io/__;!!Mih3wA!CiwC6zfeIyr6DIp0qqUZNapI4qIaSCYUAhPtwuWfaKJ_wGfhuYyZrGljygq6qYUY__bSvW_nN3mL3SY1vfTyLg$ 
https://urldefense.com/v3/__https://orcid.org/0000-0003-4010-4632__;!!Mih3wA!CiwC6zfeIyr6DIp0qqUZNapI4qIaSCYUAhPtwuWfaKJ_wGfhuYyZrGljygq6qYUY__bSvW_nN3mL3SZ8LYVr2A$