[FieldTrip] how to calculate effect size for F-test with contrasts

Alba Peris Yag├╝e alba.peris at ctb.upm.es
Thu May 5 16:31:09 CEST 2022


Dear Fieldtrip community,

I am working with human iEEG data. After conducting a cluster-corrected
dependent sample MANOVA with a contrast (code below), I have found a
significant cluster of activity from 62-90Hz from 0.19-0.56s.

I would now like to calculate the effect size of my effect. However, I am
not sure what the best approach is since I am using an F test with
contrasts.

Does anyone have any indications on how to calculate it?

Thank you very much,

Alba

cfg = [];

cfg.method           = 'ft_statistics_montecarlo';

cfg.statistic        = 'ft_statfun_depsamplesFmultivariate';

cfg.correctm         = 'cluster';

cfg.clusterthreshold = 'nonparametric_common';

cfg.clusterstatistic = 'maxsize';

cfg.latency          = [0 1];

cfg.frequency        = foi;

cfg.tail             = 1;

cfg.clustertail      = 1;

cfg.alpha            = 0.05;

cfg.clusteralpha     = 0.05;

cfg.numrandomization = 1000;

cfg.neighbours = [];

cfg.ivar = 1;

cfg.uvar = 2;

cfg.design = [ones(1,totalsubj) ones(1,totalsubj).*2
ones(1,totalsubj).*3;[1:totalsubj] [1:totalsubj] [1:totalsubj]];

cfg.contrastcoefs    = [1 1 -2];

stat_result = ft_freqstatistics(cfg,A,B,C);
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20220505/ca0c6883/attachment.htm>


More information about the fieldtrip mailing list