Hi,

I recently conducted a linear model to compare the effect of variate 1 (code as 0 and 1, i.e two group comparison) adjusted for variate 2 as a covariate, how should I set the matrix, "variate 1+variate 2,0+1+0+0+0,50" or "variate 1+variate 2,0+1+1+0+0,50"? And how to interpret the linear model results since the meta-analysis inherently involves a comparison of the functional alteration between patient vs healthy control?

Thanks for answering all these questions. I really appreciate any help!

Best,
Chengyu

Hello,i am also confused of the linear model. When i used this to compare two groups ,i got no results.I set sdm_stable a group variable to seperate the two groups(A for-1 and B for 1).But My model is ''modelline = mi_lm ,0+0+0+0+0,50,,1''    I just do not understant the 0 or 1 in this part.Have you solved this ploblem?

Hi Chengyu,

To set the linear model with variable1 adjusted by variable2 the command should be LinearModel = mi_lm var1+var2,0+1+0+0+0,50,,1.

The results indicate if there is a correlation (positive of negative) of that variable with the main results. Eg. if i have a Z value positve in a certain area when studying adult sample, that mean that area is stronger activated (in comparison of patients with healthy controls) when the sample are adults.

Kind Regards,
Lydia

Hi,

You should specify one group as 0 and the other 1 (instead of -1 and 1). When selecting the variables in your model, be sure you put  1 for the hypothesis you want to test.

Kind regards

Thank you so much! That works!