I think the gamma/rstd what you want. doug
On 11/07/2013 08:14 AM, 陈骥 wrote:
Dear freesurfer experts
In qdec I use the DODS degsin to obtain the thickness difference
between my two groups with age ,gender as nuisance.Then I want to get the effect size map instead of report the significance level. In freesurfer wiki I found that we can obtain the effect size map via gamma/sqrt(gammavar*Npilot), where Npilot is the number subjects in the pilot study.While I have two groups,one control group with 25 subjects and one patient group with 25 subjects too,I doubt that which number I should use ,the number of patient group subjects 25 or for the whole subjects across groups 50 to repalce the Npilot in the equation? And then I look for previous relevant topic in the freesurfer mail list, found that doug suggested that we can get the effect size just use gamma/rstd, I tried for that but found completely different result between I use the equation of gamma/sqrt(gammavar*Npilot), which equation should I use for I want to get the Cohen'd effect size map? Thank you in advance for your helps! Ji Chen
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer